]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[cleanup] Misc
[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 = ('ios', 'android', 'web', 'tv', '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 VERIFIED = enum.auto()
296
297
298 class YoutubeBaseInfoExtractor(InfoExtractor):
299 """Provide base functions for Youtube extractors"""
300
301 _RESERVED_NAMES = (
302 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
303 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
304 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
305 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
306
307 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
308
309 # _NETRC_MACHINE = 'youtube'
310
311 # If True it will raise an error if no login info is provided
312 _LOGIN_REQUIRED = False
313
314 _INVIDIOUS_SITES = (
315 # invidious-redirect websites
316 r'(?:www\.)?redirect\.invidious\.io',
317 r'(?:(?:www|dev)\.)?invidio\.us',
318 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
319 r'(?:www\.)?invidious\.pussthecat\.org',
320 r'(?:www\.)?invidious\.zee\.li',
321 r'(?:www\.)?invidious\.ethibox\.fr',
322 r'(?:www\.)?iv\.ggtyler\.dev',
323 r'(?:www\.)?inv\.vern\.i2p',
324 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
325 r'(?:www\.)?inv\.riverside\.rocks',
326 r'(?:www\.)?invidious\.silur\.me',
327 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
328 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
329 r'(?:www\.)?invidious\.slipfox\.xyz',
330 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
331 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
332 r'(?:www\.)?invidious\.tiekoetter\.com',
333 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
334 r'(?:www\.)?invidious\.nerdvpn\.de',
335 r'(?:www\.)?invidious\.weblibre\.org',
336 r'(?:www\.)?inv\.odyssey346\.dev',
337 r'(?:www\.)?invidious\.dhusch\.de',
338 r'(?:www\.)?iv\.melmac\.space',
339 r'(?:www\.)?watch\.thekitty\.zone',
340 r'(?:www\.)?invidious\.privacydev\.net',
341 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
342 r'(?:www\.)?invidious\.drivet\.xyz',
343 r'(?:www\.)?vid\.priv\.au',
344 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
345 r'(?:www\.)?inv\.vern\.cc',
346 r'(?:www\.)?invidious\.esmailelbob\.xyz',
347 r'(?:www\.)?invidious\.sethforprivacy\.com',
348 r'(?:www\.)?yt\.oelrichsgarcia\.de',
349 r'(?:www\.)?yt\.artemislena\.eu',
350 r'(?:www\.)?invidious\.flokinet\.to',
351 r'(?:www\.)?invidious\.baczek\.me',
352 r'(?:www\.)?y\.com\.sb',
353 r'(?:www\.)?invidious\.epicsite\.xyz',
354 r'(?:www\.)?invidious\.lidarshield\.cloud',
355 r'(?:www\.)?yt\.funami\.tech',
356 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
357 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
358 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
359 # youtube-dl invidious instances list
360 r'(?:(?:www|no)\.)?invidiou\.sh',
361 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
362 r'(?:www\.)?invidious\.kabi\.tk',
363 r'(?:www\.)?invidious\.mastodon\.host',
364 r'(?:www\.)?invidious\.zapashcanon\.fr',
365 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
366 r'(?:www\.)?invidious\.tinfoil-hat\.net',
367 r'(?:www\.)?invidious\.himiko\.cloud',
368 r'(?:www\.)?invidious\.reallyancient\.tech',
369 r'(?:www\.)?invidious\.tube',
370 r'(?:www\.)?invidiou\.site',
371 r'(?:www\.)?invidious\.site',
372 r'(?:www\.)?invidious\.xyz',
373 r'(?:www\.)?invidious\.nixnet\.xyz',
374 r'(?:www\.)?invidious\.048596\.xyz',
375 r'(?:www\.)?invidious\.drycat\.fr',
376 r'(?:www\.)?inv\.skyn3t\.in',
377 r'(?:www\.)?tube\.poal\.co',
378 r'(?:www\.)?tube\.connect\.cafe',
379 r'(?:www\.)?vid\.wxzm\.sx',
380 r'(?:www\.)?vid\.mint\.lgbt',
381 r'(?:www\.)?vid\.puffyan\.us',
382 r'(?:www\.)?yewtu\.be',
383 r'(?:www\.)?yt\.elukerio\.org',
384 r'(?:www\.)?yt\.lelux\.fi',
385 r'(?:www\.)?invidious\.ggc-project\.de',
386 r'(?:www\.)?yt\.maisputain\.ovh',
387 r'(?:www\.)?ytprivate\.com',
388 r'(?:www\.)?invidious\.13ad\.de',
389 r'(?:www\.)?invidious\.toot\.koeln',
390 r'(?:www\.)?invidious\.fdn\.fr',
391 r'(?:www\.)?watch\.nettohikari\.com',
392 r'(?:www\.)?invidious\.namazso\.eu',
393 r'(?:www\.)?invidious\.silkky\.cloud',
394 r'(?:www\.)?invidious\.exonip\.de',
395 r'(?:www\.)?invidious\.riverside\.rocks',
396 r'(?:www\.)?invidious\.blamefran\.net',
397 r'(?:www\.)?invidious\.moomoo\.de',
398 r'(?:www\.)?ytb\.trom\.tf',
399 r'(?:www\.)?yt\.cyberhost\.uk',
400 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
401 r'(?:www\.)?qklhadlycap4cnod\.onion',
402 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
403 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
404 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
405 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
406 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
407 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
408 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
409 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
410 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
411 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
412 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
413 r'(?:www\.)?piped\.kavin\.rocks',
414 r'(?:www\.)?piped\.tokhmi\.xyz',
415 r'(?:www\.)?piped\.syncpundit\.io',
416 r'(?:www\.)?piped\.mha\.fi',
417 r'(?:www\.)?watch\.whatever\.social',
418 r'(?:www\.)?piped\.garudalinux\.org',
419 r'(?:www\.)?piped\.rivo\.lol',
420 r'(?:www\.)?piped-libre\.kavin\.rocks',
421 r'(?:www\.)?yt\.jae\.fi',
422 r'(?:www\.)?piped\.mint\.lgbt',
423 r'(?:www\.)?il\.ax',
424 r'(?:www\.)?piped\.esmailelbob\.xyz',
425 r'(?:www\.)?piped\.projectsegfau\.lt',
426 r'(?:www\.)?piped\.privacydev\.net',
427 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
428 r'(?:www\.)?piped\.smnz\.de',
429 r'(?:www\.)?piped\.adminforge\.de',
430 r'(?:www\.)?watch\.whatevertinfoil\.de',
431 r'(?:www\.)?piped\.qdi\.fi',
432 r'(?:www\.)?piped\.video',
433 r'(?:www\.)?piped\.aeong\.one',
434 r'(?:www\.)?piped\.moomoo\.me',
435 r'(?:www\.)?piped\.chauvet\.pro',
436 r'(?:www\.)?watch\.leptons\.xyz',
437 r'(?:www\.)?pd\.vern\.cc',
438 r'(?:www\.)?piped\.hostux\.net',
439 r'(?:www\.)?piped\.lunar\.icu',
440 # Hyperpipe instances from https://hyperpipe.codeberg.page/
441 r'(?:www\.)?hyperpipe\.surge\.sh',
442 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
443 r'(?:www\.)?listen\.whatever\.social',
444 r'(?:www\.)?music\.adminforge\.de',
445 )
446
447 # extracted from account/account_menu ep
448 # XXX: These are the supported YouTube UI and API languages,
449 # which is slightly different from languages supported for translation in YouTube studio
450 _SUPPORTED_LANG_CODES = [
451 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
452 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
453 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
454 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
455 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
456 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
457 ]
458
459 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
460
461 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
462 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
463
464 def ucid_or_none(self, ucid):
465 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
466
467 def handle_or_none(self, handle):
468 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
469
470 def handle_from_url(self, url):
471 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
472 url, 'channel handle', default=None)
473
474 def ucid_from_url(self, url):
475 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
476 url, 'channel id', default=None)
477
478 @functools.cached_property
479 def _preferred_lang(self):
480 """
481 Returns a language code supported by YouTube for the user preferred language.
482 Returns None if no preferred language set.
483 """
484 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
485 if not preferred_lang:
486 return
487 if preferred_lang not in self._SUPPORTED_LANG_CODES:
488 raise ExtractorError(
489 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
490 expected=True)
491 elif preferred_lang != 'en':
492 self.report_warning(
493 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
494 return preferred_lang
495
496 def _initialize_consent(self):
497 cookies = self._get_cookies('https://www.youtube.com/')
498 if cookies.get('__Secure-3PSID'):
499 return
500 consent_id = None
501 consent = cookies.get('CONSENT')
502 if consent:
503 if 'YES' in consent.value:
504 return
505 consent_id = self._search_regex(
506 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
507 if not consent_id:
508 consent_id = random.randint(100, 999)
509 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
510
511 def _initialize_pref(self):
512 cookies = self._get_cookies('https://www.youtube.com/')
513 pref_cookie = cookies.get('PREF')
514 pref = {}
515 if pref_cookie:
516 try:
517 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
518 except ValueError:
519 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
520 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
521 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
522
523 def _real_initialize(self):
524 self._initialize_pref()
525 self._initialize_consent()
526 self._check_login_required()
527
528 def _check_login_required(self):
529 if self._LOGIN_REQUIRED and not self._cookies_passed:
530 self.raise_login_required('Login details are needed to download this content', method='cookies')
531
532 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
533 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
534
535 def _get_default_ytcfg(self, client='web'):
536 return copy.deepcopy(INNERTUBE_CLIENTS[client])
537
538 def _get_innertube_host(self, client='web'):
539 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
540
541 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
542 # try_get but with fallback to default ytcfg client values when present
543 _func = lambda y: try_get(y, getter, expected_type)
544 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
545
546 def _extract_client_name(self, ytcfg, default_client='web'):
547 return self._ytcfg_get_safe(
548 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
549 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
550
551 def _extract_client_version(self, ytcfg, default_client='web'):
552 return self._ytcfg_get_safe(
553 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
554 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
555
556 def _select_api_hostname(self, req_api_hostname, default_client=None):
557 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
558 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
559
560 def _extract_api_key(self, ytcfg=None, default_client='web'):
561 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
562
563 def _extract_context(self, ytcfg=None, default_client='web'):
564 context = get_first(
565 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
566 # Enforce language and tz for extraction
567 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
568 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
569 return context
570
571 _SAPISID = None
572
573 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
574 time_now = round(time.time())
575 if self._SAPISID is None:
576 yt_cookies = self._get_cookies('https://www.youtube.com')
577 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
578 # See: https://github.com/yt-dlp/yt-dlp/issues/393
579 sapisid_cookie = dict_get(
580 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
581 if sapisid_cookie and sapisid_cookie.value:
582 self._SAPISID = sapisid_cookie.value
583 self.write_debug('Extracted SAPISID cookie')
584 # SAPISID cookie is required if not already present
585 if not yt_cookies.get('SAPISID'):
586 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
587 self._set_cookie(
588 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
589 else:
590 self._SAPISID = False
591 if not self._SAPISID:
592 return None
593 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
594 sapisidhash = hashlib.sha1(
595 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
596 return f'SAPISIDHASH {time_now}_{sapisidhash}'
597
598 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
599 note='Downloading API JSON', errnote='Unable to download API page',
600 context=None, api_key=None, api_hostname=None, default_client='web'):
601
602 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
603 data.update(query)
604 real_headers = self.generate_api_headers(default_client=default_client)
605 real_headers.update({'content-type': 'application/json'})
606 if headers:
607 real_headers.update(headers)
608 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
609 or api_key or self._extract_api_key(default_client=default_client))
610 return self._download_json(
611 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
612 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
613 data=json.dumps(data).encode('utf8'), headers=real_headers,
614 query={'key': api_key, 'prettyPrint': 'false'})
615
616 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
617 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
618
619 @staticmethod
620 def _extract_session_index(*data):
621 """
622 Index of current account in account list.
623 See: https://github.com/yt-dlp/yt-dlp/pull/519
624 """
625 for ytcfg in data:
626 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
627 if session_index is not None:
628 return session_index
629
630 # Deprecated?
631 def _extract_identity_token(self, ytcfg=None, webpage=None):
632 if ytcfg:
633 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
634 if token:
635 return token
636 if webpage:
637 return self._search_regex(
638 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
639 'identity token', default=None, fatal=False)
640
641 @staticmethod
642 def _extract_account_syncid(*args):
643 """
644 Extract syncId required to download private playlists of secondary channels
645 @params response and/or ytcfg
646 """
647 for data in args:
648 # ytcfg includes channel_syncid if on secondary channel
649 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
650 if delegated_sid:
651 return delegated_sid
652 sync_ids = (try_get(
653 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
654 lambda x: x['DATASYNC_ID']), str) or '').split('||')
655 if len(sync_ids) >= 2 and sync_ids[1]:
656 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
657 # and just "user_syncid||" for primary channel. We only want the channel_syncid
658 return sync_ids[0]
659
660 @staticmethod
661 def _extract_visitor_data(*args):
662 """
663 Extracts visitorData from an API response or ytcfg
664 Appears to be used to track session state
665 """
666 return get_first(
667 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
668 expected_type=str)
669
670 @functools.cached_property
671 def is_authenticated(self):
672 return bool(self._generate_sapisidhash_header())
673
674 def extract_ytcfg(self, video_id, webpage):
675 if not webpage:
676 return {}
677 return self._parse_json(
678 self._search_regex(
679 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
680 default='{}'), video_id, fatal=False) or {}
681
682 def generate_api_headers(
683 self, *, ytcfg=None, account_syncid=None, session_index=None,
684 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
685
686 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
687 headers = {
688 'X-YouTube-Client-Name': str(
689 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
690 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
691 'Origin': origin,
692 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
693 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
694 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
695 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
696 }
697 if session_index is None:
698 session_index = self._extract_session_index(ytcfg)
699 if account_syncid or session_index is not None:
700 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
701
702 auth = self._generate_sapisidhash_header(origin)
703 if auth is not None:
704 headers['Authorization'] = auth
705 headers['X-Origin'] = origin
706 return filter_dict(headers)
707
708 def _download_ytcfg(self, client, video_id):
709 url = {
710 'web': 'https://www.youtube.com',
711 'web_music': 'https://music.youtube.com',
712 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
713 }.get(client)
714 if not url:
715 return {}
716 webpage = self._download_webpage(
717 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
718 return self.extract_ytcfg(video_id, webpage) or {}
719
720 @staticmethod
721 def _build_api_continuation_query(continuation, ctp=None):
722 query = {
723 'continuation': continuation
724 }
725 # TODO: Inconsistency with clickTrackingParams.
726 # Currently we have a fixed ctp contained within context (from ytcfg)
727 # and a ctp in root query for continuation.
728 if ctp:
729 query['clickTracking'] = {'clickTrackingParams': ctp}
730 return query
731
732 @classmethod
733 def _extract_next_continuation_data(cls, renderer):
734 next_continuation = try_get(
735 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
736 lambda x: x['continuation']['reloadContinuationData']), dict)
737 if not next_continuation:
738 return
739 continuation = next_continuation.get('continuation')
740 if not continuation:
741 return
742 ctp = next_continuation.get('clickTrackingParams')
743 return cls._build_api_continuation_query(continuation, ctp)
744
745 @classmethod
746 def _extract_continuation_ep_data(cls, continuation_ep: dict):
747 if isinstance(continuation_ep, dict):
748 continuation = try_get(
749 continuation_ep, lambda x: x['continuationCommand']['token'], str)
750 if not continuation:
751 return
752 ctp = continuation_ep.get('clickTrackingParams')
753 return cls._build_api_continuation_query(continuation, ctp)
754
755 @classmethod
756 def _extract_continuation(cls, renderer):
757 next_continuation = cls._extract_next_continuation_data(renderer)
758 if next_continuation:
759 return next_continuation
760
761 return traverse_obj(renderer, (
762 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
763 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
764 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
765
766 @classmethod
767 def _extract_alerts(cls, data):
768 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
769 if not isinstance(alert_dict, dict):
770 continue
771 for alert in alert_dict.values():
772 alert_type = alert.get('type')
773 if not alert_type:
774 continue
775 message = cls._get_text(alert, 'text')
776 if message:
777 yield alert_type, message
778
779 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
780 errors, warnings = [], []
781 for alert_type, alert_message in alerts:
782 if alert_type.lower() == 'error' and fatal:
783 errors.append([alert_type, alert_message])
784 elif alert_message not in self._IGNORED_WARNINGS:
785 warnings.append([alert_type, alert_message])
786
787 for alert_type, alert_message in (warnings + errors[:-1]):
788 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
789 if errors:
790 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
791
792 def _extract_and_report_alerts(self, data, *args, **kwargs):
793 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
794
795 def _extract_badges(self, badge_list: list):
796 """
797 Extract known BadgeType's from a list of badge renderers.
798 @returns [{'type': BadgeType}]
799 """
800 icon_type_map = {
801 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
802 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
803 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
804 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
805 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
806 'CHECK': BadgeType.VERIFIED,
807 }
808
809 badge_style_map = {
810 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
811 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
812 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
813 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
814 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
815 }
816
817 label_map = {
818 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
819 'private': BadgeType.AVAILABILITY_PRIVATE,
820 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
821 'live': BadgeType.LIVE_NOW,
822 'premium': BadgeType.AVAILABILITY_PREMIUM,
823 'verified': BadgeType.VERIFIED,
824 'official artist channel': BadgeType.VERIFIED,
825 }
826
827 badges = []
828 for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
829 badge_type = (
830 icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
831 or badge_style_map.get(traverse_obj(badge, 'style'))
832 )
833 if badge_type:
834 badges.append({'type': badge_type})
835 continue
836
837 # fallback, won't work in some languages
838 label = traverse_obj(
839 badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
840 for match, label_badge_type in label_map.items():
841 if match in label.lower():
842 badges.append({'type': label_badge_type})
843 break
844
845 return badges
846
847 @staticmethod
848 def _has_badge(badges, badge_type):
849 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
850
851 @staticmethod
852 def _get_text(data, *path_list, max_runs=None):
853 for path in path_list or [None]:
854 if path is None:
855 obj = [data]
856 else:
857 obj = traverse_obj(data, path, default=[])
858 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
859 obj = [obj]
860 for item in obj:
861 text = try_get(item, lambda x: x['simpleText'], str)
862 if text:
863 return text
864 runs = try_get(item, lambda x: x['runs'], list) or []
865 if not runs and isinstance(item, list):
866 runs = item
867
868 runs = runs[:min(len(runs), max_runs or len(runs))]
869 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
870 if text:
871 return text
872
873 def _get_count(self, data, *path_list):
874 count_text = self._get_text(data, *path_list) or ''
875 count = parse_count(count_text)
876 if count is None:
877 count = str_to_int(
878 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
879 return count
880
881 @staticmethod
882 def _extract_thumbnails(data, *path_list):
883 """
884 Extract thumbnails from thumbnails dict
885 @param path_list: path list to level that contains 'thumbnails' key
886 """
887 thumbnails = []
888 for path in path_list or [()]:
889 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
890 thumbnail_url = url_or_none(thumbnail.get('url'))
891 if not thumbnail_url:
892 continue
893 # Sometimes youtube gives a wrong thumbnail URL. See:
894 # https://github.com/yt-dlp/yt-dlp/issues/233
895 # https://github.com/ytdl-org/youtube-dl/issues/28023
896 if 'maxresdefault' in thumbnail_url:
897 thumbnail_url = thumbnail_url.split('?')[0]
898 thumbnails.append({
899 'url': thumbnail_url,
900 'height': int_or_none(thumbnail.get('height')),
901 'width': int_or_none(thumbnail.get('width')),
902 })
903 return thumbnails
904
905 @staticmethod
906 def extract_relative_time(relative_time_text):
907 """
908 Extracts a relative time from string and converts to dt object
909 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
910 """
911
912 # XXX: this could be moved to a general function in utils.py
913 # The relative time text strings are roughly the same as what
914 # Javascript's Intl.RelativeTimeFormat function generates.
915 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
916 mobj = re.search(
917 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',
918 relative_time_text)
919 if mobj:
920 start = mobj.group('start')
921 if start:
922 return datetime_from_str(start)
923 try:
924 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
925 except ValueError:
926 return None
927
928 def _parse_time_text(self, text):
929 if not text:
930 return
931 dt = self.extract_relative_time(text)
932 timestamp = None
933 if isinstance(dt, datetime.datetime):
934 timestamp = calendar.timegm(dt.timetuple())
935
936 if timestamp is None:
937 timestamp = (
938 unified_timestamp(text) or unified_timestamp(
939 self._search_regex(
940 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
941 text.lower(), 'time text', default=None)))
942
943 if text and timestamp is None and self._preferred_lang in (None, 'en'):
944 self.report_warning(
945 f'Cannot parse localized time text "{text}"', only_once=True)
946 return timestamp
947
948 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
949 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
950 default_client='web'):
951 for retry in self.RetryManager():
952 try:
953 response = self._call_api(
954 ep=ep, fatal=True, headers=headers,
955 video_id=item_id, query=query, note=note,
956 context=self._extract_context(ytcfg, default_client),
957 api_key=self._extract_api_key(ytcfg, default_client),
958 api_hostname=api_hostname, default_client=default_client)
959 except ExtractorError as e:
960 if not isinstance(e.cause, network_exceptions):
961 return self._error_or_warning(e, fatal=fatal)
962 elif not isinstance(e.cause, urllib.error.HTTPError):
963 retry.error = e
964 continue
965
966 first_bytes = e.cause.read(512)
967 if not is_html(first_bytes):
968 yt_error = try_get(
969 self._parse_json(
970 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
971 lambda x: x['error']['message'], str)
972 if yt_error:
973 self._report_alerts([('ERROR', yt_error)], fatal=False)
974 # Downloading page may result in intermittent 5xx HTTP error
975 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
976 # We also want to catch all other network exceptions since errors in later pages can be troublesome
977 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
978 if e.cause.code not in (403, 429):
979 retry.error = e
980 continue
981 return self._error_or_warning(e, fatal=fatal)
982
983 try:
984 self._extract_and_report_alerts(response, only_once=True)
985 except ExtractorError as e:
986 # YouTube servers may return errors we want to retry on in a 200 OK response
987 # See: https://github.com/yt-dlp/yt-dlp/issues/839
988 if 'unknown error' in e.msg.lower():
989 retry.error = e
990 continue
991 return self._error_or_warning(e, fatal=fatal)
992 # Youtube sometimes sends incomplete data
993 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
994 if not traverse_obj(response, *variadic(check_get_keys)):
995 retry.error = ExtractorError('Incomplete data received', expected=True)
996 continue
997
998 return response
999
1000 @staticmethod
1001 def is_music_url(url):
1002 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1003
1004 def _extract_video(self, renderer):
1005 video_id = renderer.get('videoId')
1006
1007 reel_header_renderer = traverse_obj(renderer, (
1008 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1009 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1010
1011 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1012 description = self._get_text(renderer, 'descriptionSnippet')
1013
1014 duration = int_or_none(renderer.get('lengthSeconds'))
1015 if duration is None:
1016 duration = parse_duration(self._get_text(
1017 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1018 if duration is None:
1019 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1020 duration = parse_duration(self._search_regex(
1021 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1022 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1023 video_id, default=None, group='duration'))
1024
1025 channel_id = traverse_obj(
1026 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1027 expected_type=str, get_all=False)
1028 if not channel_id:
1029 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1030
1031 channel_id = self.ucid_or_none(channel_id)
1032
1033 overlay_style = traverse_obj(
1034 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1035 get_all=False, expected_type=str)
1036 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1037 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1038 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1039 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1040 expected_type=str)) or ''
1041 url = f'https://www.youtube.com/watch?v={video_id}'
1042 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1043 url = f'https://www.youtube.com/shorts/{video_id}'
1044
1045 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1046 or self._get_text(reel_header_renderer, 'timestampText') or '')
1047 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1048
1049 live_status = (
1050 'is_upcoming' if scheduled_timestamp is not None
1051 else 'was_live' if 'streamed' in time_text.lower()
1052 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1053 else None)
1054
1055 # videoInfo is a string like '50K views • 10 years ago'.
1056 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1057 view_count = (0 if 'no views' in view_count_text.lower()
1058 else self._get_count({'simpleText': view_count_text}))
1059 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1060
1061 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1062 or self._get_text(reel_header_renderer, 'channelTitleText'))
1063
1064 channel_handle = traverse_obj(renderer, (
1065 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1066 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1067 expected_type=self.handle_from_url, get_all=False)
1068 return {
1069 '_type': 'url',
1070 'ie_key': YoutubeIE.ie_key(),
1071 'id': video_id,
1072 'url': url,
1073 'title': title,
1074 'description': description,
1075 'duration': duration,
1076 'channel_id': channel_id,
1077 'channel': channel,
1078 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1079 'uploader': channel,
1080 'uploader_id': channel_handle,
1081 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1082 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1083 'timestamp': (self._parse_time_text(time_text)
1084 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1085 else None),
1086 'release_timestamp': scheduled_timestamp,
1087 'availability':
1088 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1089 else self._availability(
1090 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1091 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1092 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1093 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1094 view_count_field: view_count,
1095 'live_status': live_status,
1096 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
1097 }
1098
1099
1100 class YoutubeIE(YoutubeBaseInfoExtractor):
1101 IE_DESC = 'YouTube'
1102 _VALID_URL = r"""(?x)^
1103 (
1104 (?:https?://|//) # http(s):// or protocol-independent URL
1105 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1106 (?:www\.)?deturl\.com/www\.youtube\.com|
1107 (?:www\.)?pwnyoutube\.com|
1108 (?:www\.)?hooktube\.com|
1109 (?:www\.)?yourepeat\.com|
1110 tube\.majestyc\.net|
1111 %(invidious)s|
1112 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1113 (?:.*?\#/)? # handle anchor (#/) redirect urls
1114 (?: # the various things that can precede the ID:
1115 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1116 |(?: # or the v= param in all its forms
1117 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1118 (?:\?|\#!?) # the params delimiter ? or # or #!
1119 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1120 v=
1121 )
1122 ))
1123 |(?:
1124 youtu\.be| # just youtu.be/xxxx
1125 vid\.plus| # or vid.plus/xxxx
1126 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1127 %(invidious)s
1128 )/
1129 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1130 )
1131 )? # all until now is optional -> you can pass the naked ID
1132 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1133 (?(1).+)? # if we found the ID, everything can follow
1134 (?:\#|$)""" % {
1135 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1136 }
1137 _EMBED_REGEX = [
1138 r'''(?x)
1139 (?:
1140 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1141 data-video-url=|
1142 <embed[^>]+?src=|
1143 embedSWF\(?:\s*|
1144 <object[^>]+data=|
1145 new\s+SWFObject\(
1146 )
1147 (["\'])
1148 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1149 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1150 \1''',
1151 # https://wordpress.org/plugins/lazy-load-for-videos/
1152 r'''(?xs)
1153 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1154 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1155 ]
1156 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1157
1158 _PLAYER_INFO_RE = (
1159 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1160 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$',
1161 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1162 )
1163 _formats = {
1164 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1165 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1166 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1167 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1168 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1169 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1170 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1171 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1172 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1173 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1174 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1175 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1176 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1177 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1178 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1179 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1180 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1181 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1182
1183
1184 # 3D videos
1185 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1186 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1187 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1188 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1189 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1190 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1191 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1192
1193 # Apple HTTP Live Streaming
1194 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1195 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1196 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1197 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1198 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1199 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1200 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1201 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1202
1203 # DASH mp4 video
1204 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1205 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1206 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1207 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1208 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1209 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1210 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1211 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1212 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1213 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1214 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1215 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1216
1217 # Dash mp4 audio
1218 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1219 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1220 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1221 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1222 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1223 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1224 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1225
1226 # Dash webm
1227 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1228 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1229 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1230 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1231 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1232 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1233 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1234 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1235 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1236 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1237 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1238 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1239 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1240 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1241 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1242 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1243 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1244 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1245 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1246 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1247 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1248 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1249
1250 # Dash webm audio
1251 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1252 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1253
1254 # Dash webm audio with opus inside
1255 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1256 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1257 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1258
1259 # RTMP (unnamed)
1260 '_rtmp': {'protocol': 'rtmp'},
1261
1262 # av01 video only formats sometimes served with "unknown" codecs
1263 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1264 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1265 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1266 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1267 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1268 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1269 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1270 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1271 }
1272 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1273
1274 _GEO_BYPASS = False
1275
1276 IE_NAME = 'youtube'
1277 _TESTS = [
1278 {
1279 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1280 'info_dict': {
1281 'id': 'BaW_jenozKc',
1282 'ext': 'mp4',
1283 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1284 'channel': 'Philipp Hagemeister',
1285 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1286 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1287 'upload_date': '20121002',
1288 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1289 'categories': ['Science & Technology'],
1290 'tags': ['youtube-dl'],
1291 'duration': 10,
1292 'view_count': int,
1293 'like_count': int,
1294 'availability': 'public',
1295 'playable_in_embed': True,
1296 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1297 'live_status': 'not_live',
1298 'age_limit': 0,
1299 'start_time': 1,
1300 'end_time': 9,
1301 'comment_count': int,
1302 'channel_follower_count': int,
1303 'uploader': 'Philipp Hagemeister',
1304 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1305 'uploader_id': '@PhilippHagemeister',
1306 'heatmap': 'count:100',
1307 }
1308 },
1309 {
1310 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1311 'note': 'Embed-only video (#1746)',
1312 'info_dict': {
1313 'id': 'yZIXLfi8CZQ',
1314 'ext': 'mp4',
1315 'upload_date': '20120608',
1316 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1317 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1318 'age_limit': 18,
1319 },
1320 'skip': 'Private video',
1321 },
1322 {
1323 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1324 'note': 'Use the first video ID in the URL',
1325 'info_dict': {
1326 'id': 'BaW_jenozKc',
1327 'ext': 'mp4',
1328 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1329 'channel': 'Philipp Hagemeister',
1330 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1331 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1332 'upload_date': '20121002',
1333 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1334 'categories': ['Science & Technology'],
1335 'tags': ['youtube-dl'],
1336 'duration': 10,
1337 'view_count': int,
1338 'like_count': int,
1339 'availability': 'public',
1340 'playable_in_embed': True,
1341 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1342 'live_status': 'not_live',
1343 'age_limit': 0,
1344 'comment_count': int,
1345 'channel_follower_count': int,
1346 'uploader': 'Philipp Hagemeister',
1347 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1348 'uploader_id': '@PhilippHagemeister',
1349 'heatmap': 'count:100',
1350 },
1351 'params': {
1352 'skip_download': True,
1353 },
1354 },
1355 {
1356 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1357 'note': '256k DASH audio (format 141) via DASH manifest',
1358 'info_dict': {
1359 'id': 'a9LDPn-MO4I',
1360 'ext': 'm4a',
1361 'upload_date': '20121002',
1362 'description': '',
1363 'title': 'UHDTV TEST 8K VIDEO.mp4'
1364 },
1365 'params': {
1366 'youtube_include_dash_manifest': True,
1367 'format': '141',
1368 },
1369 'skip': 'format 141 not served anymore',
1370 },
1371 # DASH manifest with encrypted signature
1372 {
1373 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1374 'info_dict': {
1375 'id': 'IB3lcPjvWLA',
1376 'ext': 'm4a',
1377 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1378 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1379 'duration': 244,
1380 'upload_date': '20131011',
1381 'abr': 129.495,
1382 'like_count': int,
1383 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1384 'playable_in_embed': True,
1385 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1386 'view_count': int,
1387 'track': 'The Spark',
1388 'live_status': 'not_live',
1389 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1390 'channel': 'Afrojack',
1391 'tags': 'count:19',
1392 'availability': 'public',
1393 'categories': ['Music'],
1394 'age_limit': 0,
1395 'alt_title': 'The Spark',
1396 'channel_follower_count': int,
1397 'uploader': 'Afrojack',
1398 'uploader_url': 'https://www.youtube.com/@Afrojack',
1399 'uploader_id': '@Afrojack',
1400 },
1401 'params': {
1402 'youtube_include_dash_manifest': True,
1403 'format': '141/bestaudio[ext=m4a]',
1404 },
1405 },
1406 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1407 {
1408 'note': 'Embed allowed age-gate video',
1409 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1410 'info_dict': {
1411 'id': 'HtVdAasjOgU',
1412 'ext': 'mp4',
1413 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1414 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1415 'duration': 142,
1416 'upload_date': '20140605',
1417 'age_limit': 18,
1418 'categories': ['Gaming'],
1419 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1420 'availability': 'needs_auth',
1421 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1422 'like_count': int,
1423 'channel': 'The Witcher',
1424 'live_status': 'not_live',
1425 'tags': 'count:17',
1426 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1427 'playable_in_embed': True,
1428 'view_count': int,
1429 'channel_follower_count': int,
1430 'uploader': 'The Witcher',
1431 'uploader_url': 'https://www.youtube.com/@thewitcher',
1432 'uploader_id': '@thewitcher',
1433 'comment_count': int,
1434 'channel_is_verified': True,
1435 'heatmap': 'count:100',
1436 },
1437 },
1438 {
1439 'note': 'Age-gate video with embed allowed in public site',
1440 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1441 'info_dict': {
1442 'id': 'HsUATh_Nc2U',
1443 'ext': 'mp4',
1444 'title': 'Godzilla 2 (Official Video)',
1445 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1446 'upload_date': '20200408',
1447 'age_limit': 18,
1448 'availability': 'needs_auth',
1449 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1450 'channel': 'FlyingKitty',
1451 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1452 'view_count': int,
1453 'categories': ['Entertainment'],
1454 'live_status': 'not_live',
1455 'tags': ['Flyingkitty', 'godzilla 2'],
1456 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1457 'like_count': int,
1458 'duration': 177,
1459 'playable_in_embed': True,
1460 'channel_follower_count': int,
1461 'uploader': 'FlyingKitty',
1462 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1463 'uploader_id': '@FlyingKitty900',
1464 'comment_count': int,
1465 'channel_is_verified': True,
1466 },
1467 },
1468 {
1469 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1470 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1471 'info_dict': {
1472 'id': 'Tq92D6wQ1mg',
1473 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1474 'ext': 'mp4',
1475 'upload_date': '20191228',
1476 'description': 'md5:17eccca93a786d51bc67646756894066',
1477 'age_limit': 18,
1478 'like_count': int,
1479 'availability': 'needs_auth',
1480 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1481 'view_count': int,
1482 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1483 'channel': 'Projekt Melody',
1484 'live_status': 'not_live',
1485 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1486 'playable_in_embed': True,
1487 'categories': ['Entertainment'],
1488 'duration': 106,
1489 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1490 'comment_count': int,
1491 'channel_follower_count': int,
1492 'uploader': 'Projekt Melody',
1493 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1494 'uploader_id': '@ProjektMelody',
1495 },
1496 },
1497 {
1498 'note': 'Non-Agegated non-embeddable video',
1499 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1500 'info_dict': {
1501 'id': 'MeJVWBSsPAY',
1502 'ext': 'mp4',
1503 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1504 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1505 'upload_date': '20130730',
1506 'track': 'Such mich find mich',
1507 'age_limit': 0,
1508 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1509 'like_count': int,
1510 'playable_in_embed': False,
1511 'creator': 'OOMPH!',
1512 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1513 'view_count': int,
1514 'alt_title': 'Such mich find mich',
1515 'duration': 210,
1516 'channel': 'Herr Lurik',
1517 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1518 'categories': ['Music'],
1519 'availability': 'public',
1520 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1521 'live_status': 'not_live',
1522 'artist': 'OOMPH!',
1523 'channel_follower_count': int,
1524 'uploader': 'Herr Lurik',
1525 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1526 'uploader_id': '@HerrLurik',
1527 },
1528 },
1529 {
1530 'note': 'Non-bypassable age-gated video',
1531 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1532 'only_matching': True,
1533 },
1534 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1535 # YouTube Red ad is not captured for creator
1536 {
1537 'url': '__2ABJjxzNo',
1538 'info_dict': {
1539 'id': '__2ABJjxzNo',
1540 'ext': 'mp4',
1541 'duration': 266,
1542 'upload_date': '20100430',
1543 'creator': 'deadmau5',
1544 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1545 'title': 'Deadmau5 - Some Chords (HD)',
1546 'alt_title': 'Some Chords',
1547 'availability': 'public',
1548 'tags': 'count:14',
1549 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1550 'view_count': int,
1551 'live_status': 'not_live',
1552 'channel': 'deadmau5',
1553 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1554 'like_count': int,
1555 'track': 'Some Chords',
1556 'artist': 'deadmau5',
1557 'playable_in_embed': True,
1558 'age_limit': 0,
1559 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1560 'categories': ['Music'],
1561 'album': 'Some Chords',
1562 'channel_follower_count': int,
1563 'uploader': 'deadmau5',
1564 'uploader_url': 'https://www.youtube.com/@deadmau5',
1565 'uploader_id': '@deadmau5',
1566 },
1567 'expected_warnings': [
1568 'DASH manifest missing',
1569 ]
1570 },
1571 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1572 {
1573 'url': 'lqQg6PlCWgI',
1574 'info_dict': {
1575 'id': 'lqQg6PlCWgI',
1576 'ext': 'mp4',
1577 'duration': 6085,
1578 'upload_date': '20150827',
1579 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1580 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1581 'like_count': int,
1582 'release_timestamp': 1343767800,
1583 'playable_in_embed': True,
1584 'categories': ['Sports'],
1585 'release_date': '20120731',
1586 'channel': 'Olympics',
1587 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1588 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1589 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1590 'age_limit': 0,
1591 'availability': 'public',
1592 'live_status': 'was_live',
1593 'view_count': int,
1594 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1595 'channel_follower_count': int,
1596 'uploader': 'Olympics',
1597 'uploader_url': 'https://www.youtube.com/@Olympics',
1598 'uploader_id': '@Olympics',
1599 'channel_is_verified': True,
1600 },
1601 'params': {
1602 'skip_download': 'requires avconv',
1603 }
1604 },
1605 # Non-square pixels
1606 {
1607 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1608 'info_dict': {
1609 'id': '_b-2C3KPAM0',
1610 'ext': 'mp4',
1611 'stretched_ratio': 16 / 9.,
1612 'duration': 85,
1613 'upload_date': '20110310',
1614 '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',
1615 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1616 'playable_in_embed': True,
1617 'channel': '孫ᄋᄅ',
1618 'age_limit': 0,
1619 'tags': 'count:11',
1620 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1621 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1622 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1623 'view_count': int,
1624 'categories': ['People & Blogs'],
1625 'like_count': int,
1626 'live_status': 'not_live',
1627 'availability': 'unlisted',
1628 'comment_count': int,
1629 'channel_follower_count': int,
1630 'uploader': '孫ᄋᄅ',
1631 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1632 'uploader_id': '@AllenMeow',
1633 },
1634 },
1635 # url_encoded_fmt_stream_map is empty string
1636 {
1637 'url': 'qEJwOuvDf7I',
1638 'info_dict': {
1639 'id': 'qEJwOuvDf7I',
1640 'ext': 'webm',
1641 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1642 'description': '',
1643 'upload_date': '20150404',
1644 },
1645 'params': {
1646 'skip_download': 'requires avconv',
1647 },
1648 'skip': 'This live event has ended.',
1649 },
1650 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1651 {
1652 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1653 'info_dict': {
1654 'id': 'FIl7x6_3R5Y',
1655 'ext': 'webm',
1656 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1657 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1658 'duration': 220,
1659 'upload_date': '20150625',
1660 'formats': 'mincount:31',
1661 },
1662 'skip': 'not actual anymore',
1663 },
1664 # DASH manifest with segment_list
1665 {
1666 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1667 'md5': '8ce563a1d667b599d21064e982ab9e31',
1668 'info_dict': {
1669 'id': 'CsmdDsKjzN8',
1670 'ext': 'mp4',
1671 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1672 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1673 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1674 },
1675 'params': {
1676 'youtube_include_dash_manifest': True,
1677 'format': '135', # bestvideo
1678 },
1679 'skip': 'This live event has ended.',
1680 },
1681 {
1682 # Multifeed videos (multiple cameras), URL can be of any Camera
1683 # TODO: fix multifeed titles
1684 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1685 'info_dict': {
1686 'id': 'zaPI8MvL8pg',
1687 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1688 'description': 'md5:563ccbc698b39298481ca3c571169519',
1689 },
1690 'playlist': [{
1691 'info_dict': {
1692 'id': 'j5yGuxZ8lLU',
1693 'ext': 'mp4',
1694 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1695 'description': 'md5:563ccbc698b39298481ca3c571169519',
1696 'duration': 10120,
1697 'channel_follower_count': int,
1698 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1699 'availability': 'public',
1700 'playable_in_embed': True,
1701 'upload_date': '20131105',
1702 'categories': ['Gaming'],
1703 'live_status': 'was_live',
1704 'tags': 'count:24',
1705 'release_timestamp': 1383701910,
1706 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1707 'comment_count': int,
1708 'age_limit': 0,
1709 'like_count': int,
1710 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1711 'channel': 'WiiLikeToPlay',
1712 'view_count': int,
1713 'release_date': '20131106',
1714 'uploader': 'WiiLikeToPlay',
1715 'uploader_id': '@WLTP',
1716 'uploader_url': 'https://www.youtube.com/@WLTP',
1717 },
1718 }, {
1719 'info_dict': {
1720 'id': 'zaPI8MvL8pg',
1721 'ext': 'mp4',
1722 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1723 'availability': 'public',
1724 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1725 'channel': 'WiiLikeToPlay',
1726 'channel_follower_count': int,
1727 'description': 'md5:563ccbc698b39298481ca3c571169519',
1728 'duration': 10108,
1729 'age_limit': 0,
1730 'like_count': int,
1731 'tags': 'count:24',
1732 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1733 'release_timestamp': 1383701915,
1734 'comment_count': int,
1735 'upload_date': '20131105',
1736 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1737 'release_date': '20131106',
1738 'playable_in_embed': True,
1739 'live_status': 'was_live',
1740 'categories': ['Gaming'],
1741 'view_count': int,
1742 'uploader': 'WiiLikeToPlay',
1743 'uploader_id': '@WLTP',
1744 'uploader_url': 'https://www.youtube.com/@WLTP',
1745 },
1746 }, {
1747 'info_dict': {
1748 'id': 'R7r3vfO7Hao',
1749 'ext': 'mp4',
1750 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1751 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1752 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1753 'like_count': int,
1754 'availability': 'public',
1755 'playable_in_embed': True,
1756 'upload_date': '20131105',
1757 'description': 'md5:563ccbc698b39298481ca3c571169519',
1758 'channel_follower_count': int,
1759 'tags': 'count:24',
1760 'release_date': '20131106',
1761 'comment_count': int,
1762 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1763 'channel': 'WiiLikeToPlay',
1764 'categories': ['Gaming'],
1765 'release_timestamp': 1383701914,
1766 'live_status': 'was_live',
1767 'age_limit': 0,
1768 'duration': 10128,
1769 'view_count': int,
1770 'uploader': 'WiiLikeToPlay',
1771 'uploader_id': '@WLTP',
1772 'uploader_url': 'https://www.youtube.com/@WLTP',
1773 },
1774 }],
1775 'params': {'skip_download': True},
1776 },
1777 {
1778 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1779 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1780 'info_dict': {
1781 'id': 'gVfLd0zydlo',
1782 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1783 },
1784 'playlist_count': 2,
1785 'skip': 'Not multifeed anymore',
1786 },
1787 {
1788 'url': 'https://vid.plus/FlRa-iH7PGw',
1789 'only_matching': True,
1790 },
1791 {
1792 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1793 'only_matching': True,
1794 },
1795 {
1796 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1797 # Also tests cut-off URL expansion in video description (see
1798 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1799 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1800 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1801 'info_dict': {
1802 'id': 'lsguqyKfVQg',
1803 'ext': 'mp4',
1804 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1805 'alt_title': 'Dark Walk',
1806 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1807 'duration': 133,
1808 'upload_date': '20151119',
1809 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1810 'track': 'Dark Walk',
1811 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1812 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1813 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1814 'categories': ['Film & Animation'],
1815 'view_count': int,
1816 'live_status': 'not_live',
1817 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1818 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1819 'tags': 'count:13',
1820 'availability': 'public',
1821 'channel': 'IronSoulElf',
1822 'playable_in_embed': True,
1823 'like_count': int,
1824 'age_limit': 0,
1825 'channel_follower_count': int
1826 },
1827 'params': {
1828 'skip_download': True,
1829 },
1830 },
1831 {
1832 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1833 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1834 'only_matching': True,
1835 },
1836 {
1837 # Video with yt:stretch=17:0
1838 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1839 'info_dict': {
1840 'id': 'Q39EVAstoRM',
1841 'ext': 'mp4',
1842 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1843 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1844 'upload_date': '20151107',
1845 },
1846 'params': {
1847 'skip_download': True,
1848 },
1849 'skip': 'This video does not exist.',
1850 },
1851 {
1852 # Video with incomplete 'yt:stretch=16:'
1853 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1854 'only_matching': True,
1855 },
1856 {
1857 # Video licensed under Creative Commons
1858 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1859 'info_dict': {
1860 'id': 'M4gD1WSo5mA',
1861 'ext': 'mp4',
1862 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1863 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1864 'duration': 721,
1865 'upload_date': '20150128',
1866 'license': 'Creative Commons Attribution license (reuse allowed)',
1867 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1868 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1869 'like_count': int,
1870 'age_limit': 0,
1871 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1872 'channel': 'The Berkman Klein Center for Internet & Society',
1873 'availability': 'public',
1874 'view_count': int,
1875 'categories': ['Education'],
1876 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1877 'live_status': 'not_live',
1878 'playable_in_embed': True,
1879 'channel_follower_count': int,
1880 'chapters': list,
1881 'uploader': 'The Berkman Klein Center for Internet & Society',
1882 'uploader_id': '@BKCHarvard',
1883 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1884 },
1885 'params': {
1886 'skip_download': True,
1887 },
1888 },
1889 {
1890 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1891 'info_dict': {
1892 'id': 'eQcmzGIKrzg',
1893 'ext': 'mp4',
1894 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1895 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1896 'duration': 4060,
1897 'upload_date': '20151120',
1898 'license': 'Creative Commons Attribution license (reuse allowed)',
1899 'playable_in_embed': True,
1900 'tags': 'count:12',
1901 'like_count': int,
1902 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1903 'age_limit': 0,
1904 'availability': 'public',
1905 'categories': ['News & Politics'],
1906 'channel': 'Bernie Sanders',
1907 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1908 'view_count': int,
1909 'live_status': 'not_live',
1910 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1911 'comment_count': int,
1912 'channel_follower_count': int,
1913 'chapters': list,
1914 'uploader': 'Bernie Sanders',
1915 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1916 'uploader_id': '@BernieSanders',
1917 'channel_is_verified': True,
1918 'heatmap': 'count:100',
1919 },
1920 'params': {
1921 'skip_download': True,
1922 },
1923 },
1924 {
1925 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1926 'only_matching': True,
1927 },
1928 {
1929 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1930 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1931 'only_matching': True,
1932 },
1933 {
1934 # Rental video preview
1935 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1936 'info_dict': {
1937 'id': 'uGpuVWrhIzE',
1938 'ext': 'mp4',
1939 'title': 'Piku - Trailer',
1940 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1941 'upload_date': '20150811',
1942 'license': 'Standard YouTube License',
1943 },
1944 'params': {
1945 'skip_download': True,
1946 },
1947 'skip': 'This video is not available.',
1948 },
1949 {
1950 # YouTube Red video with episode data
1951 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1952 'info_dict': {
1953 'id': 'iqKdEhx-dD4',
1954 'ext': 'mp4',
1955 'title': 'Isolation - Mind Field (Ep 1)',
1956 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1957 'duration': 2085,
1958 'upload_date': '20170118',
1959 'series': 'Mind Field',
1960 'season_number': 1,
1961 'episode_number': 1,
1962 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1963 'tags': 'count:12',
1964 'view_count': int,
1965 'availability': 'public',
1966 'age_limit': 0,
1967 'channel': 'Vsauce',
1968 'episode': 'Episode 1',
1969 'categories': ['Entertainment'],
1970 'season': 'Season 1',
1971 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1972 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1973 'like_count': int,
1974 'playable_in_embed': True,
1975 'live_status': 'not_live',
1976 'channel_follower_count': int,
1977 'uploader': 'Vsauce',
1978 'uploader_url': 'https://www.youtube.com/@Vsauce',
1979 'uploader_id': '@Vsauce',
1980 'comment_count': int,
1981 'channel_is_verified': True,
1982 },
1983 'params': {
1984 'skip_download': True,
1985 },
1986 'expected_warnings': [
1987 'Skipping DASH manifest',
1988 ],
1989 },
1990 {
1991 # The following content has been identified by the YouTube community
1992 # as inappropriate or offensive to some audiences.
1993 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1994 'info_dict': {
1995 'id': '6SJNVb0GnPI',
1996 'ext': 'mp4',
1997 'title': 'Race Differences in Intelligence',
1998 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1999 'duration': 965,
2000 'upload_date': '20140124',
2001 },
2002 'params': {
2003 'skip_download': True,
2004 },
2005 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2006 },
2007 {
2008 # itag 212
2009 'url': '1t24XAntNCY',
2010 'only_matching': True,
2011 },
2012 {
2013 # geo restricted to JP
2014 'url': 'sJL6WA-aGkQ',
2015 'only_matching': True,
2016 },
2017 {
2018 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2019 'only_matching': True,
2020 },
2021 {
2022 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2023 'only_matching': True,
2024 },
2025 {
2026 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2027 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2028 'only_matching': True,
2029 },
2030 {
2031 # DRM protected
2032 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2033 'only_matching': True,
2034 },
2035 {
2036 # Video with unsupported adaptive stream type formats
2037 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2038 'info_dict': {
2039 'id': 'Z4Vy8R84T1U',
2040 'ext': 'mp4',
2041 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2042 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2043 'duration': 433,
2044 'upload_date': '20130923',
2045 'formats': 'maxcount:10',
2046 },
2047 'params': {
2048 'skip_download': True,
2049 'youtube_include_dash_manifest': False,
2050 },
2051 'skip': 'not actual anymore',
2052 },
2053 {
2054 # Youtube Music Auto-generated description
2055 # TODO: fix metadata extraction
2056 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2057 'info_dict': {
2058 'id': 'MgNrAu2pzNs',
2059 'ext': 'mp4',
2060 'title': 'Voyeur Girl',
2061 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2062 'upload_date': '20190312',
2063 'artist': 'Stephen',
2064 'track': 'Voyeur Girl',
2065 'album': 'it\'s too much love to know my dear',
2066 'release_date': '20190313',
2067 'release_year': 2019,
2068 'alt_title': 'Voyeur Girl',
2069 'view_count': int,
2070 'playable_in_embed': True,
2071 'like_count': int,
2072 'categories': ['Music'],
2073 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2074 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2075 'uploader': 'Stephen',
2076 'availability': 'public',
2077 'creator': 'Stephen',
2078 'duration': 169,
2079 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2080 'age_limit': 0,
2081 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2082 'tags': 'count:11',
2083 'live_status': 'not_live',
2084 'channel_follower_count': int
2085 },
2086 'params': {
2087 'skip_download': True,
2088 },
2089 },
2090 {
2091 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2092 'only_matching': True,
2093 },
2094 {
2095 # invalid -> valid video id redirection
2096 'url': 'DJztXj2GPfl',
2097 'info_dict': {
2098 'id': 'DJztXj2GPfk',
2099 'ext': 'mp4',
2100 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2101 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2102 'upload_date': '20090125',
2103 'artist': 'Panjabi MC',
2104 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2105 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2106 },
2107 'params': {
2108 'skip_download': True,
2109 },
2110 'skip': 'Video unavailable',
2111 },
2112 {
2113 # empty description results in an empty string
2114 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2115 'info_dict': {
2116 'id': 'x41yOUIvK2k',
2117 'ext': 'mp4',
2118 'title': 'IMG 3456',
2119 'description': '',
2120 'upload_date': '20170613',
2121 'view_count': int,
2122 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2123 'like_count': int,
2124 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2125 'tags': [],
2126 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2127 'availability': 'public',
2128 'age_limit': 0,
2129 'categories': ['Pets & Animals'],
2130 'duration': 7,
2131 'playable_in_embed': True,
2132 'live_status': 'not_live',
2133 'channel': 'l\'Or Vert asbl',
2134 'channel_follower_count': int,
2135 'uploader': 'l\'Or Vert asbl',
2136 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2137 'uploader_id': '@ElevageOrVert',
2138 },
2139 'params': {
2140 'skip_download': True,
2141 },
2142 },
2143 {
2144 # with '};' inside yt initial data (see [1])
2145 # see [2] for an example with '};' inside ytInitialPlayerResponse
2146 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2147 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2148 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2149 'info_dict': {
2150 'id': 'CHqg6qOn4no',
2151 'ext': 'mp4',
2152 'title': 'Part 77 Sort a list of simple types in c#',
2153 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2154 'upload_date': '20130831',
2155 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2156 'like_count': int,
2157 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2158 'live_status': 'not_live',
2159 'categories': ['Education'],
2160 'availability': 'public',
2161 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2162 'tags': 'count:12',
2163 'playable_in_embed': True,
2164 'age_limit': 0,
2165 'view_count': int,
2166 'duration': 522,
2167 'channel': 'kudvenkat',
2168 'comment_count': int,
2169 'channel_follower_count': int,
2170 'chapters': list,
2171 'uploader': 'kudvenkat',
2172 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2173 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2174 'channel_is_verified': True,
2175 'heatmap': 'count:100',
2176 },
2177 'params': {
2178 'skip_download': True,
2179 },
2180 },
2181 {
2182 # another example of '};' in ytInitialData
2183 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2184 'only_matching': True,
2185 },
2186 {
2187 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2188 'only_matching': True,
2189 },
2190 {
2191 # https://github.com/ytdl-org/youtube-dl/pull/28094
2192 'url': 'OtqTfy26tG0',
2193 'info_dict': {
2194 'id': 'OtqTfy26tG0',
2195 'ext': 'mp4',
2196 'title': 'Burn Out',
2197 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2198 'upload_date': '20141120',
2199 'artist': 'The Cinematic Orchestra',
2200 'track': 'Burn Out',
2201 'album': 'Every Day',
2202 'like_count': int,
2203 'live_status': 'not_live',
2204 'alt_title': 'Burn Out',
2205 'duration': 614,
2206 'age_limit': 0,
2207 'view_count': int,
2208 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2209 'creator': 'The Cinematic Orchestra',
2210 'channel': 'The Cinematic Orchestra',
2211 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2212 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2213 'availability': 'public',
2214 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2215 'categories': ['Music'],
2216 'playable_in_embed': True,
2217 'channel_follower_count': int,
2218 'uploader': 'The Cinematic Orchestra',
2219 'comment_count': int,
2220 },
2221 'params': {
2222 'skip_download': True,
2223 },
2224 },
2225 {
2226 # controversial video, only works with bpctr when authenticated with cookies
2227 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2228 'only_matching': True,
2229 },
2230 {
2231 # controversial video, requires bpctr/contentCheckOk
2232 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2233 'info_dict': {
2234 'id': 'SZJvDhaSDnc',
2235 'ext': 'mp4',
2236 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2237 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2238 'upload_date': '20140716',
2239 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2240 'duration': 170,
2241 'categories': ['News & Politics'],
2242 'view_count': int,
2243 'channel': 'CBS Mornings',
2244 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2245 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2246 'age_limit': 18,
2247 'availability': 'needs_auth',
2248 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2249 'like_count': int,
2250 'live_status': 'not_live',
2251 'playable_in_embed': True,
2252 'channel_follower_count': int,
2253 'uploader': 'CBS Mornings',
2254 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2255 'uploader_id': '@CBSMornings',
2256 'comment_count': int,
2257 'channel_is_verified': True,
2258 }
2259 },
2260 {
2261 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2262 'url': 'cBvYw8_A0vQ',
2263 'info_dict': {
2264 'id': 'cBvYw8_A0vQ',
2265 'ext': 'mp4',
2266 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2267 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2268 'upload_date': '20201120',
2269 'duration': 1456,
2270 'categories': ['Travel & Events'],
2271 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2272 'view_count': int,
2273 'channel': 'Walk around Japan',
2274 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2275 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2276 'age_limit': 0,
2277 'availability': 'public',
2278 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2279 'live_status': 'not_live',
2280 'playable_in_embed': True,
2281 'channel_follower_count': int,
2282 'uploader': 'Walk around Japan',
2283 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2284 'uploader_id': '@walkaroundjapan7124',
2285 },
2286 'params': {
2287 'skip_download': True,
2288 },
2289 }, {
2290 # Has multiple audio streams
2291 'url': 'WaOKSUlf4TM',
2292 'only_matching': True
2293 }, {
2294 # Requires Premium: has format 141 when requested using YTM url
2295 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2296 'only_matching': True
2297 }, {
2298 # multiple subtitles with same lang_code
2299 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2300 'only_matching': True,
2301 }, {
2302 # Force use android client fallback
2303 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2304 'info_dict': {
2305 'id': 'YOelRv7fMxY',
2306 'title': 'DIGGING A SECRET TUNNEL Part 1',
2307 'ext': '3gp',
2308 'upload_date': '20210624',
2309 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2310 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2311 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2312 'duration': 596,
2313 'categories': ['Entertainment'],
2314 'view_count': int,
2315 'channel': 'colinfurze',
2316 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2317 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2318 'age_limit': 0,
2319 'availability': 'public',
2320 'like_count': int,
2321 'live_status': 'not_live',
2322 'playable_in_embed': True,
2323 'channel_follower_count': int,
2324 'chapters': list,
2325 'uploader': 'colinfurze',
2326 'uploader_url': 'https://www.youtube.com/@colinfurze',
2327 'uploader_id': '@colinfurze',
2328 'comment_count': int,
2329 'channel_is_verified': True,
2330 'heatmap': 'count:100',
2331 },
2332 'params': {
2333 'format': '17', # 3gp format available on android
2334 'extractor_args': {'youtube': {'player_client': ['android']}},
2335 },
2336 },
2337 {
2338 # Skip download of additional client configs (remix client config in this case)
2339 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2340 'only_matching': True,
2341 'params': {
2342 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2343 },
2344 }, {
2345 # shorts
2346 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2347 'only_matching': True,
2348 }, {
2349 'note': 'Storyboards',
2350 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2351 'info_dict': {
2352 'id': '5KLPxDtMqe8',
2353 'ext': 'mhtml',
2354 'format_id': 'sb0',
2355 'title': 'Your Brain is Plastic',
2356 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2357 'upload_date': '20140324',
2358 'like_count': int,
2359 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2360 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2361 'view_count': int,
2362 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2363 'playable_in_embed': True,
2364 'tags': 'count:12',
2365 'availability': 'public',
2366 'channel': 'SciShow',
2367 'live_status': 'not_live',
2368 'duration': 248,
2369 'categories': ['Education'],
2370 'age_limit': 0,
2371 'channel_follower_count': int,
2372 'chapters': list,
2373 'uploader': 'SciShow',
2374 'uploader_url': 'https://www.youtube.com/@SciShow',
2375 'uploader_id': '@SciShow',
2376 'comment_count': int,
2377 'channel_is_verified': True,
2378 'heatmap': 'count:100',
2379 }, 'params': {'format': 'mhtml', 'skip_download': True}
2380 }, {
2381 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2382 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2383 'info_dict': {
2384 'id': '2NUZ8W2llS4',
2385 'ext': 'mp4',
2386 'title': 'The NP that test your phone performance 🙂',
2387 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2388 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2389 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2390 'duration': 21,
2391 'view_count': int,
2392 'age_limit': 0,
2393 'categories': ['Gaming'],
2394 'tags': 'count:23',
2395 'playable_in_embed': True,
2396 'live_status': 'not_live',
2397 'upload_date': '20220103',
2398 'like_count': int,
2399 'availability': 'public',
2400 'channel': 'Leon Nguyen',
2401 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2402 'comment_count': int,
2403 'channel_follower_count': int,
2404 'uploader': 'Leon Nguyen',
2405 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2406 'uploader_id': '@LeonNguyen',
2407 'heatmap': 'count:100',
2408 }
2409 }, {
2410 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2411 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2412 'info_dict': {
2413 'id': '2NUZ8W2llS4',
2414 'ext': 'mp4',
2415 'title': 'The NP that test your phone performance 🙂',
2416 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2417 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2418 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2419 'duration': 21,
2420 'view_count': int,
2421 'age_limit': 0,
2422 'categories': ['Gaming'],
2423 'tags': 'count:23',
2424 'playable_in_embed': True,
2425 'live_status': 'not_live',
2426 'upload_date': '20220102',
2427 'like_count': int,
2428 'availability': 'public',
2429 'channel': 'Leon Nguyen',
2430 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2431 'comment_count': int,
2432 'channel_follower_count': int,
2433 'uploader': 'Leon Nguyen',
2434 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2435 'uploader_id': '@LeonNguyen',
2436 'heatmap': 'count:100',
2437 },
2438 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2439 }, {
2440 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2441 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2442 'info_dict': {
2443 'id': 'mzZzzBU6lrM',
2444 'ext': 'mp4',
2445 'title': 'I Met GeorgeNotFound In Real Life...',
2446 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2447 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2448 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2449 'duration': 955,
2450 'view_count': int,
2451 'age_limit': 0,
2452 'categories': ['Entertainment'],
2453 'tags': 'count:26',
2454 'playable_in_embed': True,
2455 'live_status': 'not_live',
2456 'release_timestamp': 1641172509,
2457 'release_date': '20220103',
2458 'upload_date': '20220103',
2459 'like_count': int,
2460 'availability': 'public',
2461 'channel': 'Quackity',
2462 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2463 'channel_follower_count': int,
2464 'uploader': 'Quackity',
2465 'uploader_id': '@Quackity',
2466 'uploader_url': 'https://www.youtube.com/@Quackity',
2467 'comment_count': int,
2468 'channel_is_verified': True,
2469 'heatmap': 'count:100',
2470 }
2471 },
2472 { # continuous livestream. Microformat upload date should be preferred.
2473 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2474 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2475 'info_dict': {
2476 'id': 'kgx4WGK0oNU',
2477 '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}',
2478 'ext': 'mp4',
2479 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2480 'availability': 'public',
2481 'age_limit': 0,
2482 'release_timestamp': 1637975704,
2483 'upload_date': '20210619',
2484 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2485 'live_status': 'is_live',
2486 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2487 'channel': 'Abao in Tokyo',
2488 'channel_follower_count': int,
2489 'release_date': '20211127',
2490 'tags': 'count:39',
2491 'categories': ['People & Blogs'],
2492 'like_count': int,
2493 'view_count': int,
2494 'playable_in_embed': True,
2495 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2496 'concurrent_view_count': int,
2497 'uploader': 'Abao in Tokyo',
2498 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2499 'uploader_id': '@abaointokyo',
2500 },
2501 'params': {'skip_download': True}
2502 }, {
2503 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2504 'info_dict': {
2505 'id': 'tjjjtzRLHvA',
2506 'ext': 'mp4',
2507 'title': 'ハッシュタグ無し };if window.ytcsi',
2508 'upload_date': '20220323',
2509 'like_count': int,
2510 'availability': 'unlisted',
2511 'channel': 'Lesmiscore',
2512 'thumbnail': r're:^https?://.*\.jpg',
2513 'age_limit': 0,
2514 'categories': ['Music'],
2515 'view_count': int,
2516 'description': '',
2517 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2518 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2519 'live_status': 'not_live',
2520 'playable_in_embed': True,
2521 'channel_follower_count': int,
2522 'duration': 6,
2523 'tags': [],
2524 'uploader_id': '@lesmiscore',
2525 'uploader': 'Lesmiscore',
2526 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2527 }
2528 }, {
2529 # Prefer primary title+description language metadata by default
2530 # Do not prefer translated description if primary is empty
2531 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2532 'info_dict': {
2533 'id': 'el3E4MbxRqQ',
2534 'ext': 'mp4',
2535 'title': 'dlp test video 2 - primary sv no desc',
2536 'description': '',
2537 'channel': 'cole-dlp-test-acc',
2538 'tags': [],
2539 'view_count': int,
2540 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2541 'like_count': int,
2542 'playable_in_embed': True,
2543 'availability': 'unlisted',
2544 'thumbnail': r're:^https?://.*\.jpg',
2545 'age_limit': 0,
2546 'duration': 5,
2547 'live_status': 'not_live',
2548 'upload_date': '20220908',
2549 'categories': ['People & Blogs'],
2550 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2551 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2552 'uploader_id': '@coletdjnz',
2553 'uploader': 'cole-dlp-test-acc',
2554 },
2555 'params': {'skip_download': True}
2556 }, {
2557 # Extractor argument: prefer translated title+description
2558 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2559 'info_dict': {
2560 'id': 'gHKT4uU8Zng',
2561 'ext': 'mp4',
2562 'channel': 'cole-dlp-test-acc',
2563 'tags': [],
2564 'duration': 5,
2565 'live_status': 'not_live',
2566 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2567 'upload_date': '20220728',
2568 'view_count': int,
2569 'categories': ['People & Blogs'],
2570 'thumbnail': r're:^https?://.*\.jpg',
2571 'title': 'dlp test video title translated (fr)',
2572 'availability': 'public',
2573 'age_limit': 0,
2574 'description': 'dlp test video description translated (fr)',
2575 'playable_in_embed': True,
2576 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2577 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2578 'uploader_id': '@coletdjnz',
2579 'uploader': 'cole-dlp-test-acc',
2580 },
2581 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2582 'expected_warnings': [r'Preferring "fr" translated fields'],
2583 }, {
2584 'note': '6 channel audio',
2585 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2586 'only_matching': True,
2587 }, {
2588 'note': 'Multiple HLS formats with same itag',
2589 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2590 'info_dict': {
2591 'id': 'kX3nB4PpJko',
2592 'ext': 'mp4',
2593 'categories': ['Entertainment'],
2594 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2595 'live_status': 'not_live',
2596 'duration': 937,
2597 'channel_follower_count': int,
2598 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2599 'title': 'Last To Take Hand Off Jet, Keeps It!',
2600 'channel': 'MrBeast',
2601 'playable_in_embed': True,
2602 'view_count': int,
2603 'upload_date': '20221112',
2604 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2605 'age_limit': 0,
2606 'availability': 'public',
2607 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2608 'like_count': int,
2609 'tags': [],
2610 'uploader': 'MrBeast',
2611 'uploader_url': 'https://www.youtube.com/@MrBeast',
2612 'uploader_id': '@MrBeast',
2613 'comment_count': int,
2614 'channel_is_verified': True,
2615 'heatmap': 'count:100',
2616 },
2617 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2618 }, {
2619 'note': 'Audio formats with Dynamic Range Compression',
2620 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2621 'info_dict': {
2622 'id': 'Tq92D6wQ1mg',
2623 'ext': 'webm',
2624 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2625 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2626 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2627 'channel_follower_count': int,
2628 'description': 'md5:17eccca93a786d51bc67646756894066',
2629 'upload_date': '20191228',
2630 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2631 'playable_in_embed': True,
2632 'like_count': int,
2633 'categories': ['Entertainment'],
2634 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2635 'age_limit': 18,
2636 'channel': 'Projekt Melody',
2637 'view_count': int,
2638 'availability': 'needs_auth',
2639 'comment_count': int,
2640 'live_status': 'not_live',
2641 'duration': 106,
2642 'uploader': 'Projekt Melody',
2643 'uploader_id': '@ProjektMelody',
2644 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2645 },
2646 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2647 },
2648 {
2649 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2650 'info_dict': {
2651 'id': 'qVv6vCqciTM',
2652 'ext': 'mp4',
2653 'age_limit': 0,
2654 'comment_count': int,
2655 'chapters': 'count:13',
2656 'upload_date': '20221223',
2657 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2658 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2659 'like_count': int,
2660 'release_date': '20221223',
2661 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2662 '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】',
2663 'view_count': int,
2664 'playable_in_embed': True,
2665 'duration': 4438,
2666 'availability': 'public',
2667 'channel_follower_count': int,
2668 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2669 'categories': ['Entertainment'],
2670 'live_status': 'was_live',
2671 'release_timestamp': 1671793345,
2672 'channel': 'さなちゃんねる',
2673 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2674 'uploader': 'さなちゃんねる',
2675 'uploader_url': 'https://www.youtube.com/@sana_natori',
2676 'uploader_id': '@sana_natori',
2677 'channel_is_verified': True,
2678 'heatmap': 'count:100',
2679 },
2680 },
2681 {
2682 # Fallbacks when webpage and web client is unavailable
2683 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2684 'info_dict': {
2685 'id': 'wSSmNUl9Snw',
2686 'ext': 'mp4',
2687 # 'categories': ['Science & Technology'],
2688 'view_count': int,
2689 'chapters': 'count:2',
2690 'channel': 'Scott Manley',
2691 'like_count': int,
2692 'age_limit': 0,
2693 # 'availability': 'public',
2694 'channel_follower_count': int,
2695 'live_status': 'not_live',
2696 'upload_date': '20170831',
2697 'duration': 682,
2698 'tags': 'count:8',
2699 'uploader_url': 'https://www.youtube.com/@scottmanley',
2700 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2701 'uploader': 'Scott Manley',
2702 'uploader_id': '@scottmanley',
2703 'title': 'The Computer Hack That Saved Apollo 14',
2704 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2705 'thumbnail': r're:^https?://.*\.webp',
2706 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2707 'playable_in_embed': True,
2708 'comment_count': int,
2709 'channel_is_verified': True,
2710 'heatmap': 'count:100',
2711 },
2712 'params': {
2713 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2714 },
2715 },
2716 ]
2717
2718 _WEBPAGE_TESTS = [
2719 # YouTube <object> embed
2720 {
2721 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2722 'md5': '873c81d308b979f0e23ee7e620b312a3',
2723 'info_dict': {
2724 'id': 'msN87y-iEx0',
2725 'ext': 'mp4',
2726 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2727 'upload_date': '20080526',
2728 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2729 'age_limit': 0,
2730 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2731 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2732 'playable_in_embed': True,
2733 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2734 'like_count': int,
2735 'comment_count': int,
2736 'channel': 'Christopher Sykes',
2737 'live_status': 'not_live',
2738 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2739 'availability': 'public',
2740 'duration': 195,
2741 'view_count': int,
2742 'categories': ['Science & Technology'],
2743 'channel_follower_count': int,
2744 'uploader': 'Christopher Sykes',
2745 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2746 'uploader_id': '@ChristopherSykesDocumentaries',
2747 'heatmap': 'count:100',
2748 },
2749 'params': {
2750 'skip_download': True,
2751 }
2752 },
2753 ]
2754
2755 @classmethod
2756 def suitable(cls, url):
2757 from ..utils import parse_qs
2758
2759 qs = parse_qs(url)
2760 if qs.get('list', [None])[0]:
2761 return False
2762 return super().suitable(url)
2763
2764 def __init__(self, *args, **kwargs):
2765 super().__init__(*args, **kwargs)
2766 self._code_cache = {}
2767 self._player_cache = {}
2768
2769 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2770 lock = threading.Lock()
2771 start_time = time.time()
2772 formats = [f for f in formats if f.get('is_from_start')]
2773
2774 def refetch_manifest(format_id, delay):
2775 nonlocal formats, start_time, is_live
2776 if time.time() <= start_time + delay:
2777 return
2778
2779 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2780 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2781 microformats = traverse_obj(
2782 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2783 expected_type=dict)
2784 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2785 is_live = live_status == 'is_live'
2786 start_time = time.time()
2787
2788 def mpd_feed(format_id, delay):
2789 """
2790 @returns (manifest_url, manifest_stream_number, is_live) or None
2791 """
2792 for retry in self.RetryManager(fatal=False):
2793 with lock:
2794 refetch_manifest(format_id, delay)
2795
2796 f = next((f for f in formats if f['format_id'] == format_id), None)
2797 if not f:
2798 if not is_live:
2799 retry.error = f'{video_id}: Video is no longer live'
2800 else:
2801 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2802 continue
2803 return f['manifest_url'], f['manifest_stream_number'], is_live
2804 return None
2805
2806 for f in formats:
2807 f['is_live'] = is_live
2808 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2809 live_start_time, mpd_feed, not is_live and f.copy())
2810 if is_live:
2811 f['fragments'] = gen
2812 f['protocol'] = 'http_dash_segments_generator'
2813 else:
2814 f['fragments'] = LazyList(gen({}))
2815 del f['is_from_start']
2816
2817 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2818 FETCH_SPAN, MAX_DURATION = 5, 432000
2819
2820 mpd_url, stream_number, is_live = None, None, True
2821
2822 begin_index = 0
2823 download_start_time = ctx.get('start') or time.time()
2824
2825 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2826 if lack_early_segments:
2827 self.report_warning(bug_reports_message(
2828 'Starting download from the last 120 hours of the live stream since '
2829 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2830 lack_early_segments = True
2831
2832 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2833 fragments, fragment_base_url = None, None
2834
2835 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2836 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2837 # Obtain from MPD's maximum seq value
2838 old_mpd_url = mpd_url
2839 last_error = ctx.pop('last_error', None)
2840 expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2841 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2842 or (mpd_url, stream_number, False))
2843 if not refresh_sequence:
2844 if expire_fast and not is_live:
2845 return False, last_seq
2846 elif old_mpd_url == mpd_url:
2847 return True, last_seq
2848 if manifestless_orig_fmt:
2849 fmt_info = manifestless_orig_fmt
2850 else:
2851 try:
2852 fmts, _ = self._extract_mpd_formats_and_subtitles(
2853 mpd_url, None, note=False, errnote=False, fatal=False)
2854 except ExtractorError:
2855 fmts = None
2856 if not fmts:
2857 no_fragment_score += 2
2858 return False, last_seq
2859 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2860 fragments = fmt_info['fragments']
2861 fragment_base_url = fmt_info['fragment_base_url']
2862 assert fragment_base_url
2863
2864 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2865 return True, _last_seq
2866
2867 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2868 while is_live:
2869 fetch_time = time.time()
2870 if no_fragment_score > 30:
2871 return
2872 if last_segment_url:
2873 # Obtain from "X-Head-Seqnum" header value from each segment
2874 try:
2875 urlh = self._request_webpage(
2876 last_segment_url, None, note=False, errnote=False, fatal=False)
2877 except ExtractorError:
2878 urlh = None
2879 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2880 if last_seq is None:
2881 no_fragment_score += 2
2882 last_segment_url = None
2883 continue
2884 else:
2885 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2886 no_fragment_score += 2
2887 if not should_continue:
2888 continue
2889
2890 if known_idx > last_seq:
2891 last_segment_url = None
2892 continue
2893
2894 last_seq += 1
2895
2896 if begin_index < 0 and known_idx < 0:
2897 # skip from the start when it's negative value
2898 known_idx = last_seq + begin_index
2899 if lack_early_segments:
2900 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2901 try:
2902 for idx in range(known_idx, last_seq):
2903 # do not update sequence here or you'll get skipped some part of it
2904 should_continue, _ = _extract_sequence_from_mpd(False, False)
2905 if not should_continue:
2906 known_idx = idx - 1
2907 raise ExtractorError('breaking out of outer loop')
2908 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2909 yield {
2910 'url': last_segment_url,
2911 'fragment_count': last_seq,
2912 }
2913 if known_idx == last_seq:
2914 no_fragment_score += 5
2915 else:
2916 no_fragment_score = 0
2917 known_idx = last_seq
2918 except ExtractorError:
2919 continue
2920
2921 if manifestless_orig_fmt:
2922 # Stop at the first iteration if running for post-live manifestless;
2923 # fragment count no longer increase since it starts
2924 break
2925
2926 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2927
2928 def _extract_player_url(self, *ytcfgs, webpage=None):
2929 player_url = traverse_obj(
2930 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2931 get_all=False, expected_type=str)
2932 if not player_url:
2933 return
2934 return urljoin('https://www.youtube.com', player_url)
2935
2936 def _download_player_url(self, video_id, fatal=False):
2937 res = self._download_webpage(
2938 'https://www.youtube.com/iframe_api',
2939 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2940 if res:
2941 player_version = self._search_regex(
2942 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2943 if player_version:
2944 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2945
2946 def _signature_cache_id(self, example_sig):
2947 """ Return a string representation of a signature """
2948 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2949
2950 @classmethod
2951 def _extract_player_info(cls, player_url):
2952 for player_re in cls._PLAYER_INFO_RE:
2953 id_m = re.search(player_re, player_url)
2954 if id_m:
2955 break
2956 else:
2957 raise ExtractorError('Cannot identify player %r' % player_url)
2958 return id_m.group('id')
2959
2960 def _load_player(self, video_id, player_url, fatal=True):
2961 player_id = self._extract_player_info(player_url)
2962 if player_id not in self._code_cache:
2963 code = self._download_webpage(
2964 player_url, video_id, fatal=fatal,
2965 note='Downloading player ' + player_id,
2966 errnote='Download of %s failed' % player_url)
2967 if code:
2968 self._code_cache[player_id] = code
2969 return self._code_cache.get(player_id)
2970
2971 def _extract_signature_function(self, video_id, player_url, example_sig):
2972 player_id = self._extract_player_info(player_url)
2973
2974 # Read from filesystem cache
2975 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2976 assert os.path.basename(func_id) == func_id
2977
2978 self.write_debug(f'Extracting signature function {func_id}')
2979 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2980
2981 if not cache_spec:
2982 code = self._load_player(video_id, player_url)
2983 if code:
2984 res = self._parse_sig_js(code)
2985 test_string = ''.join(map(chr, range(len(example_sig))))
2986 cache_spec = [ord(c) for c in res(test_string)]
2987 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2988
2989 return lambda s: ''.join(s[i] for i in cache_spec)
2990
2991 def _print_sig_code(self, func, example_sig):
2992 if not self.get_param('youtube_print_sig_code'):
2993 return
2994
2995 def gen_sig_code(idxs):
2996 def _genslice(start, end, step):
2997 starts = '' if start == 0 else str(start)
2998 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2999 steps = '' if step == 1 else (':%d' % step)
3000 return f's[{starts}{ends}{steps}]'
3001
3002 step = None
3003 # Quelch pyflakes warnings - start will be set when step is set
3004 start = '(Never used)'
3005 for i, prev in zip(idxs[1:], idxs[:-1]):
3006 if step is not None:
3007 if i - prev == step:
3008 continue
3009 yield _genslice(start, prev, step)
3010 step = None
3011 continue
3012 if i - prev in [-1, 1]:
3013 step = i - prev
3014 start = prev
3015 continue
3016 else:
3017 yield 's[%d]' % prev
3018 if step is None:
3019 yield 's[%d]' % i
3020 else:
3021 yield _genslice(start, i, step)
3022
3023 test_string = ''.join(map(chr, range(len(example_sig))))
3024 cache_res = func(test_string)
3025 cache_spec = [ord(c) for c in cache_res]
3026 expr_code = ' + '.join(gen_sig_code(cache_spec))
3027 signature_id_tuple = '(%s)' % (
3028 ', '.join(str(len(p)) for p in example_sig.split('.')))
3029 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3030 ' return %s\n') % (signature_id_tuple, expr_code)
3031 self.to_screen('Extracted signature function:\n' + code)
3032
3033 def _parse_sig_js(self, jscode):
3034 funcname = self._search_regex(
3035 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3036 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3037 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3038 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3039 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}\(a,\d+\))?',
3040 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3041 # Obsolete patterns
3042 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3043 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3044 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3045 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3046 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3047 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3048 jscode, 'Initial JS player signature function name', group='sig')
3049
3050 jsi = JSInterpreter(jscode)
3051 initial_function = jsi.extract_function(funcname)
3052 return lambda s: initial_function([s])
3053
3054 def _cached(self, func, *cache_id):
3055 def inner(*args, **kwargs):
3056 if cache_id not in self._player_cache:
3057 try:
3058 self._player_cache[cache_id] = func(*args, **kwargs)
3059 except ExtractorError as e:
3060 self._player_cache[cache_id] = e
3061 except Exception as e:
3062 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3063
3064 ret = self._player_cache[cache_id]
3065 if isinstance(ret, Exception):
3066 raise ret
3067 return ret
3068 return inner
3069
3070 def _decrypt_signature(self, s, video_id, player_url):
3071 """Turn the encrypted s field into a working signature"""
3072 extract_sig = self._cached(
3073 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3074 func = extract_sig(video_id, player_url, s)
3075 self._print_sig_code(func, s)
3076 return func(s)
3077
3078 def _decrypt_nsig(self, s, video_id, player_url):
3079 """Turn the encrypted n field into a working signature"""
3080 if player_url is None:
3081 raise ExtractorError('Cannot decrypt nsig without player_url')
3082 player_url = urljoin('https://www.youtube.com', player_url)
3083
3084 try:
3085 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3086 except ExtractorError as e:
3087 raise ExtractorError('Unable to extract nsig function code', cause=e)
3088 if self.get_param('youtube_print_sig_code'):
3089 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3090
3091 try:
3092 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3093 ret = extract_nsig(jsi, func_code)(s)
3094 except JSInterpreter.Exception as e:
3095 try:
3096 jsi = PhantomJSwrapper(self, timeout=5000)
3097 except ExtractorError:
3098 raise e
3099 self.report_warning(
3100 f'Native nsig extraction failed: Trying with PhantomJS\n'
3101 f' n = {s} ; player = {player_url}', video_id)
3102 self.write_debug(e, only_once=True)
3103
3104 args, func_body = func_code
3105 ret = jsi.execute(
3106 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3107 video_id=video_id, note='Executing signature code').strip()
3108
3109 self.write_debug(f'Decrypted nsig {s} => {ret}')
3110 return ret
3111
3112 def _extract_n_function_name(self, jscode):
3113 funcname, idx = self._search_regex(
3114 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3115 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3116 if not idx:
3117 return funcname
3118
3119 return json.loads(js_to_json(self._search_regex(
3120 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3121 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3122
3123 def _extract_n_function_code(self, video_id, player_url):
3124 player_id = self._extract_player_info(player_url)
3125 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3126 jscode = func_code or self._load_player(video_id, player_url)
3127 jsi = JSInterpreter(jscode)
3128
3129 if func_code:
3130 return jsi, player_id, func_code
3131
3132 func_name = self._extract_n_function_name(jscode)
3133
3134 # For redundancy
3135 func_code = self._search_regex(
3136 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3137 # NB: The end of the regex is intentionally kept strict
3138 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3139 jscode, 'nsig function', group=('var', 'code'), default=None)
3140 if func_code:
3141 func_code = ([func_code[0]], func_code[1])
3142 else:
3143 self.write_debug('Extracting nsig function with jsinterp')
3144 func_code = jsi.extract_function_code(func_name)
3145
3146 self.cache.store('youtube-nsig', player_id, func_code)
3147 return jsi, player_id, func_code
3148
3149 def _extract_n_function_from_code(self, jsi, func_code):
3150 func = jsi.extract_function_from_code(*func_code)
3151
3152 def extract_nsig(s):
3153 try:
3154 ret = func([s])
3155 except JSInterpreter.Exception:
3156 raise
3157 except Exception as e:
3158 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3159
3160 if ret.startswith('enhanced_except_'):
3161 raise JSInterpreter.Exception('Signature function returned an exception')
3162 return ret
3163
3164 return extract_nsig
3165
3166 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3167 """
3168 Extract signatureTimestamp (sts)
3169 Required to tell API what sig/player version is in use.
3170 """
3171 sts = None
3172 if isinstance(ytcfg, dict):
3173 sts = int_or_none(ytcfg.get('STS'))
3174
3175 if not sts:
3176 # Attempt to extract from player
3177 if player_url is None:
3178 error_msg = 'Cannot extract signature timestamp without player_url.'
3179 if fatal:
3180 raise ExtractorError(error_msg)
3181 self.report_warning(error_msg)
3182 return
3183 code = self._load_player(video_id, player_url, fatal=fatal)
3184 if code:
3185 sts = int_or_none(self._search_regex(
3186 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3187 'JS player signature timestamp', group='sts', fatal=fatal))
3188 return sts
3189
3190 def _mark_watched(self, video_id, player_responses):
3191 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3192 label = 'fully ' if is_full else ''
3193 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3194 expected_type=url_or_none)
3195 if not url:
3196 self.report_warning(f'Unable to mark {label}watched')
3197 return
3198 parsed_url = urllib.parse.urlparse(url)
3199 qs = urllib.parse.parse_qs(parsed_url.query)
3200
3201 # cpn generation algorithm is reverse engineered from base.js.
3202 # In fact it works even with dummy cpn.
3203 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3204 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3205
3206 # # more consistent results setting it to right before the end
3207 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3208
3209 qs.update({
3210 'ver': ['2'],
3211 'cpn': [cpn],
3212 'cmt': video_length,
3213 'el': 'detailpage', # otherwise defaults to "shorts"
3214 })
3215
3216 if is_full:
3217 # these seem to mark watchtime "history" in the real world
3218 # they're required, so send in a single value
3219 qs.update({
3220 'st': 0,
3221 'et': video_length,
3222 })
3223
3224 url = urllib.parse.urlunparse(
3225 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3226
3227 self._download_webpage(
3228 url, video_id, f'Marking {label}watched',
3229 'Unable to mark watched', fatal=False)
3230
3231 @classmethod
3232 def _extract_from_webpage(cls, url, webpage):
3233 # Invidious Instances
3234 # https://github.com/yt-dlp/yt-dlp/issues/195
3235 # https://github.com/iv-org/invidious/pull/1730
3236 mobj = re.search(
3237 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3238 webpage)
3239 if mobj:
3240 yield cls.url_result(mobj.group('url'), cls)
3241 raise cls.StopExtraction()
3242
3243 yield from super()._extract_from_webpage(url, webpage)
3244
3245 # lazyYT YouTube embed
3246 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3247 yield cls.url_result(unescapeHTML(id_), cls, id_)
3248
3249 # Wordpress "YouTube Video Importer" plugin
3250 for m in re.findall(r'''(?x)<div[^>]+
3251 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3252 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3253 yield cls.url_result(m[-1], cls, m[-1])
3254
3255 @classmethod
3256 def extract_id(cls, url):
3257 video_id = cls.get_temp_id(url)
3258 if not video_id:
3259 raise ExtractorError(f'Invalid URL: {url}')
3260 return video_id
3261
3262 def _extract_chapters_from_json(self, data, duration):
3263 chapter_list = traverse_obj(
3264 data, (
3265 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3266 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3267 ), expected_type=list)
3268
3269 return self._extract_chapters_helper(
3270 chapter_list,
3271 start_function=lambda chapter: float_or_none(
3272 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3273 title_function=lambda chapter: traverse_obj(
3274 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3275 duration=duration)
3276
3277 def _extract_chapters_from_engagement_panel(self, data, duration):
3278 content_list = traverse_obj(
3279 data,
3280 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3281 expected_type=list)
3282 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3283 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3284
3285 return next(filter(None, (
3286 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3287 chapter_time, chapter_title, duration)
3288 for contents in content_list)), [])
3289
3290 def _extract_heatmap_from_player_overlay(self, data):
3291 content_list = traverse_obj(data, (
3292 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer', 'decoratedPlayerBarRenderer', 'playerBar',
3293 'multiMarkersPlayerBarRenderer', 'markersMap', ..., 'value', 'heatmap', 'heatmapRenderer', 'heatMarkers', {list}))
3294 return next(filter(None, (
3295 traverse_obj(contents, (..., 'heatMarkerRenderer', {
3296 'start_time': ('timeRangeStartMillis', {functools.partial(float_or_none, scale=1000)}),
3297 'end_time': {lambda x: (x['timeRangeStartMillis'] + x['markerDurationMillis']) / 1000},
3298 'value': ('heatMarkerIntensityScoreNormalized', {float_or_none}),
3299 })) for contents in content_list)), None)
3300
3301 def _extract_comment(self, comment_renderer, parent=None):
3302 comment_id = comment_renderer.get('commentId')
3303 if not comment_id:
3304 return
3305
3306 info = {
3307 'id': comment_id,
3308 'text': self._get_text(comment_renderer, 'contentText'),
3309 'like_count': self._get_count(comment_renderer, 'voteCount'),
3310 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3311 'author': self._get_text(comment_renderer, 'authorText'),
3312 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3313 'parent': parent or 'root',
3314 }
3315
3316 # Timestamp is an estimate calculated from the current time and time_text
3317 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3318 timestamp = self._parse_time_text(time_text)
3319
3320 info.update({
3321 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3322 '_time_text': time_text,
3323 'timestamp': timestamp,
3324 })
3325
3326 info['author_url'] = urljoin(
3327 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3328 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3329 expected_type=str, get_all=False))
3330
3331 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3332 if author_is_uploader is not None:
3333 info['author_is_uploader'] = author_is_uploader
3334
3335 comment_abr = traverse_obj(
3336 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3337 if comment_abr is not None:
3338 info['is_favorited'] = 'creatorHeart' in comment_abr
3339
3340 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3341 if self._has_badge(badges, BadgeType.VERIFIED):
3342 info['author_is_verified'] = True
3343
3344 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3345 if is_pinned:
3346 info['is_pinned'] = True
3347
3348 return info
3349
3350 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3351
3352 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3353
3354 def extract_header(contents):
3355 _continuation = None
3356 for content in contents:
3357 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3358 expected_comment_count = self._get_count(
3359 comments_header_renderer, 'countText', 'commentsCount')
3360
3361 if expected_comment_count is not None:
3362 tracker['est_total'] = expected_comment_count
3363 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3364 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3365
3366 sort_menu_item = try_get(
3367 comments_header_renderer,
3368 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3369 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3370
3371 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3372 if not _continuation:
3373 continue
3374
3375 sort_text = str_or_none(sort_menu_item.get('title'))
3376 if not sort_text:
3377 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3378 self.to_screen('Sorting comments by %s' % sort_text.lower())
3379 break
3380 return _continuation
3381
3382 def extract_thread(contents):
3383 if not parent:
3384 tracker['current_page_thread'] = 0
3385 for content in contents:
3386 if not parent and tracker['total_parent_comments'] >= max_parents:
3387 yield
3388 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3389 comment_renderer = get_first(
3390 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3391 expected_type=dict, default={})
3392
3393 comment = self._extract_comment(comment_renderer, parent)
3394 if not comment:
3395 continue
3396 comment_id = comment['id']
3397 if comment.get('is_pinned'):
3398 tracker['pinned_comment_ids'].add(comment_id)
3399 # Sometimes YouTube may break and give us infinite looping comments.
3400 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3401 if comment_id in tracker['seen_comment_ids']:
3402 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3403 # Pinned comments may appear a second time in newest first sort
3404 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3405 continue
3406 self.report_warning(
3407 'Detected YouTube comments looping. Stopping comment extraction '
3408 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3409 yield
3410 else:
3411 tracker['seen_comment_ids'].add(comment['id'])
3412
3413 tracker['running_total'] += 1
3414 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3415 yield comment
3416
3417 # Attempt to get the replies
3418 comment_replies_renderer = try_get(
3419 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3420
3421 if comment_replies_renderer:
3422 tracker['current_page_thread'] += 1
3423 comment_entries_iter = self._comment_entries(
3424 comment_replies_renderer, ytcfg, video_id,
3425 parent=comment.get('id'), tracker=tracker)
3426 yield from itertools.islice(comment_entries_iter, min(
3427 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3428
3429 # Keeps track of counts across recursive calls
3430 if not tracker:
3431 tracker = dict(
3432 running_total=0,
3433 est_total=None,
3434 current_page_thread=0,
3435 total_parent_comments=0,
3436 total_reply_comments=0,
3437 seen_comment_ids=set(),
3438 pinned_comment_ids=set()
3439 )
3440
3441 # TODO: Deprecated
3442 # YouTube comments have a max depth of 2
3443 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3444 if max_depth:
3445 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3446 'Set max replies in the max-comments extractor argument instead')
3447 if max_depth == 1 and parent:
3448 return
3449
3450 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3451 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3452
3453 continuation = self._extract_continuation(root_continuation_data)
3454
3455 response = None
3456 is_forced_continuation = False
3457 is_first_continuation = parent is None
3458 if is_first_continuation and not continuation:
3459 # Sometimes you can get comments by generating the continuation yourself,
3460 # even if YouTube initially reports them being disabled - e.g. stories comments.
3461 # Note: if the comment section is actually disabled, YouTube may return a response with
3462 # required check_get_keys missing. So we will disable that check initially in this case.
3463 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3464 is_forced_continuation = True
3465
3466 continuation_items_path = (
3467 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3468 for page_num in itertools.count(0):
3469 if not continuation:
3470 break
3471 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3472 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3473 if page_num == 0:
3474 if is_first_continuation:
3475 note_prefix = 'Downloading comment section API JSON'
3476 else:
3477 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3478 tracker['current_page_thread'], comment_prog_str)
3479 else:
3480 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3481 ' ' if parent else '', ' replies' if parent else '',
3482 page_num, comment_prog_str)
3483
3484 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3485 # Ignore check if YouTube says the comment count is 0.
3486 check_get_keys = None
3487 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3488 check_get_keys = [[*continuation_items_path, ..., (
3489 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3490 try:
3491 response = self._extract_response(
3492 item_id=None, query=continuation,
3493 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3494 check_get_keys=check_get_keys)
3495 except ExtractorError as e:
3496 # Ignore incomplete data error for replies if retries didn't work.
3497 # This is to allow any other parent comments and comment threads to be downloaded.
3498 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3499 if 'incomplete data' in str(e).lower() and parent:
3500 if self.get_param('ignoreerrors') in (True, 'only_download'):
3501 self.report_warning(
3502 'Received incomplete data for a comment reply thread and retrying did not help. '
3503 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3504 return
3505 else:
3506 raise ExtractorError(
3507 'Incomplete data received for comment reply thread. '
3508 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3509 expected=True)
3510 raise
3511 is_forced_continuation = False
3512 continuation = None
3513 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3514 if is_first_continuation:
3515 continuation = extract_header(continuation_items)
3516 is_first_continuation = False
3517 if continuation:
3518 break
3519 continue
3520
3521 for entry in extract_thread(continuation_items):
3522 if not entry:
3523 return
3524 yield entry
3525 continuation = self._extract_continuation({'contents': continuation_items})
3526 if continuation:
3527 break
3528
3529 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3530 if message and not parent and tracker['running_total'] == 0:
3531 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3532 raise self.CommentsDisabled
3533
3534 @staticmethod
3535 def _generate_comment_continuation(video_id):
3536 """
3537 Generates initial comment section continuation token from given video id
3538 """
3539 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3540 return base64.b64encode(token.encode()).decode()
3541
3542 def _get_comments(self, ytcfg, video_id, contents, webpage):
3543 """Entry for comment extraction"""
3544 def _real_comment_extract(contents):
3545 renderer = next((
3546 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3547 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3548 yield from self._comment_entries(renderer, ytcfg, video_id)
3549
3550 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3551 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3552
3553 @staticmethod
3554 def _get_checkok_params():
3555 return {'contentCheckOk': True, 'racyCheckOk': True}
3556
3557 @classmethod
3558 def _generate_player_context(cls, sts=None):
3559 context = {
3560 'html5Preference': 'HTML5_PREF_WANTS',
3561 }
3562 if sts is not None:
3563 context['signatureTimestamp'] = sts
3564 return {
3565 'playbackContext': {
3566 'contentPlaybackContext': context
3567 },
3568 **cls._get_checkok_params()
3569 }
3570
3571 @staticmethod
3572 def _is_agegated(player_response):
3573 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3574 return True
3575
3576 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3577 AGE_GATE_REASONS = (
3578 'confirm your age', 'age-restricted', 'inappropriate', # reason
3579 'age_verification_required', 'age_check_required', # status
3580 )
3581 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3582
3583 @staticmethod
3584 def _is_unplayable(player_response):
3585 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3586
3587 _PLAYER_PARAMS = 'CgIQBg=='
3588
3589 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3590
3591 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3592 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3593 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3594 headers = self.generate_api_headers(
3595 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3596
3597 yt_query = {
3598 'videoId': video_id,
3599 }
3600 if _split_innertube_client(client)[0] == 'android':
3601 yt_query['params'] = self._PLAYER_PARAMS
3602
3603 yt_query.update(self._generate_player_context(sts))
3604 return self._extract_response(
3605 item_id=video_id, ep='player', query=yt_query,
3606 ytcfg=player_ytcfg, headers=headers, fatal=True,
3607 default_client=client,
3608 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3609 ) or None
3610
3611 def _get_requested_clients(self, url, smuggled_data):
3612 requested_clients = []
3613 default = ['ios', 'android', 'web']
3614 allowed_clients = sorted(
3615 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3616 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3617 for client in self._configuration_arg('player_client'):
3618 if client in allowed_clients:
3619 requested_clients.append(client)
3620 elif client == 'default':
3621 requested_clients.extend(default)
3622 elif client == 'all':
3623 requested_clients.extend(allowed_clients)
3624 else:
3625 self.report_warning(f'Skipping unsupported client {client}')
3626 if not requested_clients:
3627 requested_clients = default
3628
3629 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3630 requested_clients.extend(
3631 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3632
3633 return orderedSet(requested_clients)
3634
3635 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3636 initial_pr = None
3637 if webpage:
3638 initial_pr = self._search_json(
3639 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3640
3641 all_clients = set(clients)
3642 clients = clients[::-1]
3643 prs = []
3644
3645 def append_client(*client_names):
3646 """ Append the first client name that exists but not already used """
3647 for client_name in client_names:
3648 actual_client = _split_innertube_client(client_name)[0]
3649 if actual_client in INNERTUBE_CLIENTS:
3650 if actual_client not in all_clients:
3651 clients.append(client_name)
3652 all_clients.add(actual_client)
3653 return
3654
3655 # Android player_response does not have microFormats which are needed for
3656 # extraction of some data. So we return the initial_pr with formats
3657 # stripped out even if not requested by the user
3658 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3659 if initial_pr:
3660 pr = dict(initial_pr)
3661 pr['streamingData'] = None
3662 prs.append(pr)
3663
3664 last_error = None
3665 tried_iframe_fallback = False
3666 player_url = None
3667 while clients:
3668 client, base_client, variant = _split_innertube_client(clients.pop())
3669 player_ytcfg = master_ytcfg if client == 'web' else {}
3670 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3671 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3672
3673 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3674 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3675 if 'js' in self._configuration_arg('player_skip'):
3676 require_js_player = False
3677 player_url = None
3678
3679 if not player_url and not tried_iframe_fallback and require_js_player:
3680 player_url = self._download_player_url(video_id)
3681 tried_iframe_fallback = True
3682
3683 try:
3684 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3685 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3686 except ExtractorError as e:
3687 if last_error:
3688 self.report_warning(last_error)
3689 last_error = e
3690 continue
3691
3692 if pr:
3693 # YouTube may return a different video player response than expected.
3694 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3695 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3696 if pr_video_id and pr_video_id != video_id:
3697 self.report_warning(
3698 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3699 else:
3700 # Save client name for introspection later
3701 name = short_client_name(client)
3702 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3703 sd[STREAMING_DATA_CLIENT_NAME] = name
3704 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3705 f[STREAMING_DATA_CLIENT_NAME] = name
3706 prs.append(pr)
3707
3708 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3709 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3710 append_client(f'{base_client}_creator')
3711 elif self._is_agegated(pr):
3712 if variant == 'tv_embedded':
3713 append_client(f'{base_client}_embedded')
3714 elif not variant:
3715 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3716
3717 if last_error:
3718 if not len(prs):
3719 raise last_error
3720 self.report_warning(last_error)
3721 return prs, player_url
3722
3723 def _needs_live_processing(self, live_status, duration):
3724 if (live_status == 'is_live' and self.get_param('live_from_start')
3725 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3726 return live_status
3727
3728 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3729 CHUNK_SIZE = 10 << 20
3730 itags, stream_ids = collections.defaultdict(set), []
3731 itag_qualities, res_qualities = {}, {0: None}
3732 q = qualities([
3733 # Normally tiny is the smallest video-only formats. But
3734 # audio-only formats with unknown quality may get tagged as tiny
3735 'tiny',
3736 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3737 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3738 ])
3739 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3740 format_types = self._configuration_arg('formats')
3741 all_formats = 'duplicate' in format_types
3742 if self._configuration_arg('include_duplicate_formats'):
3743 all_formats = True
3744 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3745 'Use formats=duplicate extractor argument instead')
3746
3747 def build_fragments(f):
3748 return LazyList({
3749 'url': update_url_query(f['url'], {
3750 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3751 })
3752 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3753
3754 for fmt in streaming_formats:
3755 if fmt.get('targetDurationSec'):
3756 continue
3757
3758 itag = str_or_none(fmt.get('itag'))
3759 audio_track = fmt.get('audioTrack') or {}
3760 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3761 if not all_formats:
3762 if stream_id in stream_ids:
3763 continue
3764
3765 quality = fmt.get('quality')
3766 height = int_or_none(fmt.get('height'))
3767 if quality == 'tiny' or not quality:
3768 quality = fmt.get('audioQuality', '').lower() or quality
3769 # The 3gp format (17) in android client has a quality of "small",
3770 # but is actually worse than other formats
3771 if itag == '17':
3772 quality = 'tiny'
3773 if quality:
3774 if itag:
3775 itag_qualities[itag] = quality
3776 if height:
3777 res_qualities[height] = quality
3778 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3779 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3780 # number of fragment that would subsequently requested with (`&sq=N`)
3781 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3782 continue
3783
3784 fmt_url = fmt.get('url')
3785 if not fmt_url:
3786 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3787 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3788 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3789 if not all((sc, fmt_url, player_url, encrypted_sig)):
3790 continue
3791 try:
3792 fmt_url += '&%s=%s' % (
3793 traverse_obj(sc, ('sp', -1)) or 'signature',
3794 self._decrypt_signature(encrypted_sig, video_id, player_url)
3795 )
3796 except ExtractorError as e:
3797 self.report_warning('Signature extraction failed: Some formats may be missing',
3798 video_id=video_id, only_once=True)
3799 self.write_debug(e, only_once=True)
3800 continue
3801
3802 query = parse_qs(fmt_url)
3803 throttled = False
3804 if query.get('n'):
3805 try:
3806 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3807 fmt_url = update_url_query(fmt_url, {
3808 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3809 })
3810 except ExtractorError as e:
3811 phantomjs_hint = ''
3812 if isinstance(e, JSInterpreter.Exception):
3813 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3814 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3815 if player_url:
3816 self.report_warning(
3817 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3818 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3819 self.write_debug(e, only_once=True)
3820 else:
3821 self.report_warning(
3822 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3823 video_id=video_id, only_once=True)
3824 throttled = True
3825
3826 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3827 language_preference = (
3828 10 if audio_track.get('audioIsDefault') and 10
3829 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3830 else -1)
3831 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3832 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3833 # Make sure to avoid false positives with small duration differences.
3834 # E.g. __2ABJjxzNo, ySuUZEjARPY
3835 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3836 if is_damaged:
3837 self.report_warning(
3838 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3839
3840 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3841 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3842 fps = int_or_none(fmt.get('fps')) or 0
3843 dct = {
3844 'asr': int_or_none(fmt.get('audioSampleRate')),
3845 'filesize': int_or_none(fmt.get('contentLength')),
3846 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3847 'format_note': join_nonempty(
3848 join_nonempty(audio_track.get('displayName'),
3849 language_preference > 0 and ' (default)', delim=''),
3850 name, fmt.get('isDrc') and 'DRC',
3851 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3852 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3853 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3854 (self.get_param('verbose') or all_formats) and client_name,
3855 delim=', '),
3856 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3857 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3858 + (100 if 'Premium' in name else 0)),
3859 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3860 'audio_channels': fmt.get('audioChannels'),
3861 'height': height,
3862 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3863 'has_drm': bool(fmt.get('drmFamilies')),
3864 'tbr': tbr,
3865 'url': fmt_url,
3866 'width': int_or_none(fmt.get('width')),
3867 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3868 'desc' if language_preference < -1 else '') or None,
3869 'language_preference': language_preference,
3870 # Strictly de-prioritize damaged and 3gp formats
3871 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3872 }
3873 mime_mobj = re.match(
3874 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3875 if mime_mobj:
3876 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3877 dct.update(parse_codecs(mime_mobj.group(2)))
3878 if itag:
3879 itags[itag].add(('https', dct.get('language')))
3880 stream_ids.append(stream_id)
3881 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3882 if single_stream and dct.get('ext'):
3883 dct['container'] = dct['ext'] + '_dash'
3884
3885 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3886 yield {
3887 **dct,
3888 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3889 'protocol': 'http_dash_segments',
3890 'fragments': build_fragments(dct),
3891 }
3892 if all_formats or 'dashy' not in format_types:
3893 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3894 yield dct
3895
3896 needs_live_processing = self._needs_live_processing(live_status, duration)
3897 skip_bad_formats = 'incomplete' not in format_types
3898 if self._configuration_arg('include_incomplete_formats'):
3899 skip_bad_formats = False
3900 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3901 'Use formats=incomplete extractor argument instead')
3902
3903 skip_manifests = set(self._configuration_arg('skip'))
3904 if (not self.get_param('youtube_include_hls_manifest', True)
3905 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3906 or needs_live_processing and skip_bad_formats):
3907 skip_manifests.add('hls')
3908
3909 if not self.get_param('youtube_include_dash_manifest', True):
3910 skip_manifests.add('dash')
3911 if self._configuration_arg('include_live_dash'):
3912 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3913 'Use formats=incomplete extractor argument instead')
3914 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3915 skip_manifests.add('dash')
3916
3917 def process_manifest_format(f, proto, client_name, itag):
3918 key = (proto, f.get('language'))
3919 if not all_formats and key in itags[itag]:
3920 return False
3921 itags[itag].add(key)
3922
3923 if itag and all_formats:
3924 f['format_id'] = f'{itag}-{proto}'
3925 elif any(p != proto for p, _ in itags[itag]):
3926 f['format_id'] = f'{itag}-{proto}'
3927 elif itag:
3928 f['format_id'] = itag
3929
3930 if itag in ('616', '235'):
3931 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
3932 f['source_preference'] = (f.get('source_preference') or -1) + 100
3933
3934 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3935 if f['quality'] == -1 and f.get('height'):
3936 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3937 if self.get_param('verbose') or all_formats:
3938 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3939 if f.get('fps') and f['fps'] <= 1:
3940 del f['fps']
3941 return True
3942
3943 subtitles = {}
3944 for sd in streaming_data:
3945 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3946
3947 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3948 if hls_manifest_url:
3949 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3950 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3951 subtitles = self._merge_subtitles(subs, subtitles)
3952 for f in fmts:
3953 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3954 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3955 yield f
3956
3957 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3958 if dash_manifest_url:
3959 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3960 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3961 for f in formats:
3962 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3963 f['filesize'] = int_or_none(self._search_regex(
3964 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3965 if needs_live_processing:
3966 f['is_from_start'] = True
3967
3968 yield f
3969 yield subtitles
3970
3971 def _extract_storyboard(self, player_responses, duration):
3972 spec = get_first(
3973 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3974 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3975 if not base_url:
3976 return
3977 L = len(spec) - 1
3978 for i, args in enumerate(spec):
3979 args = args.split('#')
3980 counts = list(map(int_or_none, args[:5]))
3981 if len(args) != 8 or not all(counts):
3982 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3983 continue
3984 width, height, frame_count, cols, rows = counts
3985 N, sigh = args[6:]
3986
3987 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3988 fragment_count = frame_count / (cols * rows)
3989 fragment_duration = duration / fragment_count
3990 yield {
3991 'format_id': f'sb{i}',
3992 'format_note': 'storyboard',
3993 'ext': 'mhtml',
3994 'protocol': 'mhtml',
3995 'acodec': 'none',
3996 'vcodec': 'none',
3997 'url': url,
3998 'width': width,
3999 'height': height,
4000 'fps': frame_count / duration,
4001 'rows': rows,
4002 'columns': cols,
4003 'fragments': [{
4004 'url': url.replace('$M', str(j)),
4005 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4006 } for j in range(math.ceil(fragment_count))],
4007 }
4008
4009 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4010 webpage = None
4011 if 'webpage' not in self._configuration_arg('player_skip'):
4012 query = {'bpctr': '9999999999', 'has_verified': '1'}
4013 webpage = self._download_webpage(
4014 webpage_url, video_id, fatal=False, query=query)
4015
4016 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4017
4018 player_responses, player_url = self._extract_player_responses(
4019 self._get_requested_clients(url, smuggled_data),
4020 video_id, webpage, master_ytcfg, smuggled_data)
4021
4022 return webpage, master_ytcfg, player_responses, player_url
4023
4024 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4025 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4026 is_live = get_first(video_details, 'isLive')
4027 if is_live is None:
4028 is_live = get_first(live_broadcast_details, 'isLiveNow')
4029 live_content = get_first(video_details, 'isLiveContent')
4030 is_upcoming = get_first(video_details, 'isUpcoming')
4031 post_live = get_first(video_details, 'isPostLiveDvr')
4032 live_status = ('post_live' if post_live
4033 else 'is_live' if is_live
4034 else 'is_upcoming' if is_upcoming
4035 else 'was_live' if live_content
4036 else 'not_live' if False in (is_live, live_content)
4037 else None)
4038 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4039 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4040
4041 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4042
4043 def _real_extract(self, url):
4044 url, smuggled_data = unsmuggle_url(url, {})
4045 video_id = self._match_id(url)
4046
4047 base_url = self.http_scheme() + '//www.youtube.com/'
4048 webpage_url = base_url + 'watch?v=' + video_id
4049
4050 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4051
4052 playability_statuses = traverse_obj(
4053 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4054
4055 trailer_video_id = get_first(
4056 playability_statuses,
4057 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4058 expected_type=str)
4059 if trailer_video_id:
4060 return self.url_result(
4061 trailer_video_id, self.ie_key(), trailer_video_id)
4062
4063 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4064 if webpage else (lambda x: None))
4065
4066 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4067 microformats = traverse_obj(
4068 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4069 expected_type=dict)
4070
4071 translated_title = self._get_text(microformats, (..., 'title'))
4072 video_title = (self._preferred_lang and translated_title
4073 or get_first(video_details, 'title') # primary
4074 or translated_title
4075 or search_meta(['og:title', 'twitter:title', 'title']))
4076 translated_description = self._get_text(microformats, (..., 'description'))
4077 original_description = get_first(video_details, 'shortDescription')
4078 video_description = (
4079 self._preferred_lang and translated_description
4080 # If original description is blank, it will be an empty string.
4081 # Do not prefer translated description in this case.
4082 or original_description if original_description is not None else translated_description)
4083
4084 multifeed_metadata_list = get_first(
4085 player_responses,
4086 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4087 expected_type=str)
4088 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4089 if self.get_param('noplaylist'):
4090 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4091 else:
4092 entries = []
4093 feed_ids = []
4094 for feed in multifeed_metadata_list.split(','):
4095 # Unquote should take place before split on comma (,) since textual
4096 # fields may contain comma as well (see
4097 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4098 feed_data = urllib.parse.parse_qs(
4099 urllib.parse.unquote_plus(feed))
4100
4101 def feed_entry(name):
4102 return try_get(
4103 feed_data, lambda x: x[name][0], str)
4104
4105 feed_id = feed_entry('id')
4106 if not feed_id:
4107 continue
4108 feed_title = feed_entry('title')
4109 title = video_title
4110 if feed_title:
4111 title += ' (%s)' % feed_title
4112 entries.append({
4113 '_type': 'url_transparent',
4114 'ie_key': 'Youtube',
4115 'url': smuggle_url(
4116 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4117 {'force_singlefeed': True}),
4118 'title': title,
4119 })
4120 feed_ids.append(feed_id)
4121 self.to_screen(
4122 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4123 % (', '.join(feed_ids), video_id))
4124 return self.playlist_result(
4125 entries, video_id, video_title, video_description)
4126
4127 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4128 or int_or_none(get_first(microformats, 'lengthSeconds'))
4129 or parse_duration(search_meta('duration')) or None)
4130
4131 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4132 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4133 if live_status == 'post_live':
4134 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4135
4136 if not formats:
4137 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4138 self.report_drm(video_id)
4139 pemr = get_first(
4140 playability_statuses,
4141 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4142 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4143 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4144 if subreason:
4145 if subreason == 'The uploader has not made this video available in your country.':
4146 countries = get_first(microformats, 'availableCountries')
4147 if not countries:
4148 regions_allowed = search_meta('regionsAllowed')
4149 countries = regions_allowed.split(',') if regions_allowed else None
4150 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4151 reason += f'. {subreason}'
4152 if reason:
4153 self.raise_no_formats(reason, expected=True)
4154
4155 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4156 if not keywords and webpage:
4157 keywords = [
4158 unescapeHTML(m.group('content'))
4159 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4160 for keyword in keywords:
4161 if keyword.startswith('yt:stretch='):
4162 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4163 if mobj:
4164 # NB: float is intentional for forcing float division
4165 w, h = (float(v) for v in mobj.groups())
4166 if w > 0 and h > 0:
4167 ratio = w / h
4168 for f in formats:
4169 if f.get('vcodec') != 'none':
4170 f['stretched_ratio'] = ratio
4171 break
4172 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4173 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4174 if thumbnail_url:
4175 thumbnails.append({
4176 'url': thumbnail_url,
4177 })
4178 original_thumbnails = thumbnails.copy()
4179
4180 # The best resolution thumbnails sometimes does not appear in the webpage
4181 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4182 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4183 thumbnail_names = [
4184 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4185 # in resolution, these are not the custom thumbnail. So de-prioritize them
4186 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4187 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4188 ]
4189 n_thumbnail_names = len(thumbnail_names)
4190 thumbnails.extend({
4191 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4192 video_id=video_id, name=name, ext=ext,
4193 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4194 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4195 for thumb in thumbnails:
4196 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4197 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4198 self._remove_duplicate_formats(thumbnails)
4199 self._downloader._sort_thumbnails(original_thumbnails)
4200
4201 category = get_first(microformats, 'category') or search_meta('genre')
4202 channel_id = self.ucid_or_none(str_or_none(
4203 get_first(video_details, 'channelId')
4204 or get_first(microformats, 'externalChannelId')
4205 or search_meta('channelId')))
4206 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4207
4208 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4209 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4210 if not duration and live_end_time and live_start_time:
4211 duration = live_end_time - live_start_time
4212
4213 needs_live_processing = self._needs_live_processing(live_status, duration)
4214
4215 def is_bad_format(fmt):
4216 if needs_live_processing and not fmt.get('is_from_start'):
4217 return True
4218 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4219 and fmt.get('protocol') == 'http_dash_segments'):
4220 return True
4221
4222 for fmt in filter(is_bad_format, formats):
4223 fmt['preference'] = (fmt.get('preference') or -1) - 10
4224 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4225
4226 if needs_live_processing:
4227 self._prepare_live_from_start_formats(
4228 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4229
4230 formats.extend(self._extract_storyboard(player_responses, duration))
4231
4232 channel_handle = self.handle_from_url(owner_profile_url)
4233
4234 info = {
4235 'id': video_id,
4236 'title': video_title,
4237 'formats': formats,
4238 'thumbnails': thumbnails,
4239 # The best thumbnail that we are sure exists. Prevents unnecessary
4240 # URL checking if user don't care about getting the best possible thumbnail
4241 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4242 'description': video_description,
4243 'channel_id': channel_id,
4244 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4245 'duration': duration,
4246 'view_count': int_or_none(
4247 get_first((video_details, microformats), (..., 'viewCount'))
4248 or search_meta('interactionCount')),
4249 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4250 'age_limit': 18 if (
4251 get_first(microformats, 'isFamilySafe') is False
4252 or search_meta('isFamilyFriendly') == 'false'
4253 or search_meta('og:restrictions:age') == '18+') else 0,
4254 'webpage_url': webpage_url,
4255 'categories': [category] if category else None,
4256 'tags': keywords,
4257 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4258 'live_status': live_status,
4259 'release_timestamp': live_start_time,
4260 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4261 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4262 }
4263
4264 subtitles = {}
4265 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4266 if pctr:
4267 def get_lang_code(track):
4268 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4269 or track.get('languageCode'))
4270
4271 # Converted into dicts to remove duplicates
4272 captions = {
4273 get_lang_code(sub): sub
4274 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4275 translation_languages = {
4276 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4277 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4278
4279 def process_language(container, base_url, lang_code, sub_name, query):
4280 lang_subs = container.setdefault(lang_code, [])
4281 for fmt in self._SUBTITLE_FORMATS:
4282 query.update({
4283 'fmt': fmt,
4284 })
4285 lang_subs.append({
4286 'ext': fmt,
4287 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4288 'name': sub_name,
4289 })
4290
4291 # NB: Constructing the full subtitle dictionary is slow
4292 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4293 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4294 for lang_code, caption_track in captions.items():
4295 base_url = caption_track.get('baseUrl')
4296 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4297 if not base_url:
4298 continue
4299 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4300 if caption_track.get('kind') != 'asr':
4301 if not lang_code:
4302 continue
4303 process_language(
4304 subtitles, base_url, lang_code, lang_name, {})
4305 if not caption_track.get('isTranslatable'):
4306 continue
4307 for trans_code, trans_name in translation_languages.items():
4308 if not trans_code:
4309 continue
4310 orig_trans_code = trans_code
4311 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4312 if not get_translated_subs:
4313 continue
4314 trans_code += f'-{lang_code}'
4315 trans_name += format_field(lang_name, None, ' from %s')
4316 if lang_code == f'a-{orig_trans_code}':
4317 # Set audio language based on original subtitles
4318 for f in formats:
4319 if f.get('acodec') != 'none' and not f.get('language'):
4320 f['language'] = orig_trans_code
4321 # Add an "-orig" label to the original language so that it can be distinguished.
4322 # The subs are returned without "-orig" as well for compatibility
4323 process_language(
4324 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4325 # Setting tlang=lang returns damaged subtitles.
4326 process_language(automatic_captions, base_url, trans_code, trans_name,
4327 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4328
4329 info['automatic_captions'] = automatic_captions
4330 info['subtitles'] = subtitles
4331
4332 parsed_url = urllib.parse.urlparse(url)
4333 for component in [parsed_url.fragment, parsed_url.query]:
4334 query = urllib.parse.parse_qs(component)
4335 for k, v in query.items():
4336 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4337 d_k += '_time'
4338 if d_k not in info and k in s_ks:
4339 info[d_k] = parse_duration(query[k][0])
4340
4341 # Youtube Music Auto-generated description
4342 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4343 # XXX: Causes catastrophic backtracking if description has "·"
4344 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4345 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4346 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4347 mobj = re.search(
4348 r'''(?xs)
4349 (?=(?P<track>[^\n·]+))(?P=track)·
4350 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4351 (?=(?P<album>[^\n]+))(?P=album)\n
4352 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4353 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4354 (.+?\nArtist\s*:\s*
4355 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4356 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4357 ''', video_description)
4358 if mobj:
4359 release_year = mobj.group('release_year')
4360 release_date = mobj.group('release_date')
4361 if release_date:
4362 release_date = release_date.replace('-', '')
4363 if not release_year:
4364 release_year = release_date[:4]
4365 info.update({
4366 'album': mobj.group('album'.strip()),
4367 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4368 'track': mobj.group('track').strip(),
4369 'release_date': release_date,
4370 'release_year': int_or_none(release_year),
4371 })
4372
4373 initial_data = None
4374 if webpage:
4375 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4376 if not traverse_obj(initial_data, 'contents'):
4377 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4378 initial_data = None
4379 if not initial_data:
4380 query = {'videoId': video_id}
4381 query.update(self._get_checkok_params())
4382 initial_data = self._extract_response(
4383 item_id=video_id, ep='next', fatal=False,
4384 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4385 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4386 note='Downloading initial data API JSON')
4387
4388 info['comment_count'] = traverse_obj(initial_data, (
4389 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4390 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4391 ), (
4392 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4393 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4394 ), expected_type=self._get_count, get_all=False)
4395
4396 try: # This will error if there is no livechat
4397 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4398 except (KeyError, IndexError, TypeError):
4399 pass
4400 else:
4401 info.setdefault('subtitles', {})['live_chat'] = [{
4402 # url is needed to set cookies
4403 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4404 'video_id': video_id,
4405 'ext': 'json',
4406 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4407 else 'youtube_live_chat_replay'),
4408 }]
4409
4410 if initial_data:
4411 info['chapters'] = (
4412 self._extract_chapters_from_json(initial_data, duration)
4413 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4414 or self._extract_chapters_from_description(video_description, duration)
4415 or None)
4416
4417 info['heatmap'] = self._extract_heatmap_from_player_overlay(initial_data)
4418
4419 contents = traverse_obj(
4420 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4421 expected_type=list, default=[])
4422
4423 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4424 if vpir:
4425 stl = vpir.get('superTitleLink')
4426 if stl:
4427 stl = self._get_text(stl)
4428 if try_get(
4429 vpir,
4430 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4431 info['location'] = stl
4432 else:
4433 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4434 if mobj:
4435 info.update({
4436 'series': mobj.group(1),
4437 'season_number': int(mobj.group(2)),
4438 'episode_number': int(mobj.group(3)),
4439 })
4440 for tlb in (try_get(
4441 vpir,
4442 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4443 list) or []):
4444 tbrs = variadic(
4445 traverse_obj(
4446 tlb, ('toggleButtonRenderer', ...),
4447 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4448 for tbr in tbrs:
4449 for getter, regex in [(
4450 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4451 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4452 lambda x: x['accessibility'],
4453 lambda x: x['accessibilityData']['accessibilityData'],
4454 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4455 label = (try_get(tbr, getter, dict) or {}).get('label')
4456 if label:
4457 mobj = re.match(regex, label)
4458 if mobj:
4459 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4460 break
4461 sbr_tooltip = try_get(
4462 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4463 if sbr_tooltip:
4464 like_count, dislike_count = sbr_tooltip.split(' / ')
4465 info.update({
4466 'like_count': str_to_int(like_count),
4467 'dislike_count': str_to_int(dislike_count),
4468 })
4469 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4470 if vcr:
4471 vc = self._get_count(vcr, 'viewCount')
4472 # Upcoming premieres with waiting count are treated as live here
4473 if vcr.get('isLive'):
4474 info['concurrent_view_count'] = vc
4475 elif info.get('view_count') is None:
4476 info['view_count'] = vc
4477
4478 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4479 if vsir:
4480 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4481 info.update({
4482 'channel': self._get_text(vor, 'title'),
4483 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4484
4485 if not channel_handle:
4486 channel_handle = self.handle_from_url(
4487 traverse_obj(vor, (
4488 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4489 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4490 {str}), get_all=False))
4491
4492 rows = try_get(
4493 vsir,
4494 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4495 list) or []
4496 multiple_songs = False
4497 for row in rows:
4498 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4499 multiple_songs = True
4500 break
4501 for row in rows:
4502 mrr = row.get('metadataRowRenderer') or {}
4503 mrr_title = mrr.get('title')
4504 if not mrr_title:
4505 continue
4506 mrr_title = self._get_text(mrr, 'title')
4507 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4508 if mrr_title == 'License':
4509 info['license'] = mrr_contents_text
4510 elif not multiple_songs:
4511 if mrr_title == 'Album':
4512 info['album'] = mrr_contents_text
4513 elif mrr_title == 'Artist':
4514 info['artist'] = mrr_contents_text
4515 elif mrr_title == 'Song':
4516 info['track'] = mrr_contents_text
4517 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4518 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4519 info['channel_is_verified'] = True
4520
4521 info.update({
4522 'uploader': info.get('channel'),
4523 'uploader_id': channel_handle,
4524 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4525 })
4526 # The upload date for scheduled, live and past live streams / premieres in microformats
4527 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4528 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4529 upload_date = (
4530 unified_strdate(get_first(microformats, 'uploadDate'))
4531 or unified_strdate(search_meta('uploadDate')))
4532 if not upload_date or (
4533 live_status in ('not_live', None)
4534 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4535 ):
4536 upload_date = strftime_or_none(
4537 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4538 info['upload_date'] = upload_date
4539
4540 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4541 v = info.get(s_k)
4542 if v:
4543 info[d_k] = v
4544
4545 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4546
4547 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4548 or get_first(video_details, 'isPrivate', expected_type=bool))
4549
4550 info['availability'] = (
4551 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4552 else self._availability(
4553 is_private=is_private,
4554 needs_premium=(
4555 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4556 or False if initial_data and is_private is not None else None),
4557 needs_subscription=(
4558 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4559 or False if initial_data and is_private is not None else None),
4560 needs_auth=info['age_limit'] >= 18,
4561 is_unlisted=None if is_private is None else (
4562 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4563 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4564
4565 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4566
4567 self.mark_watched(video_id, player_responses)
4568
4569 return info
4570
4571
4572 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4573 @staticmethod
4574 def passthrough_smuggled_data(func):
4575 def _smuggle(info, smuggled_data):
4576 if info.get('_type') not in ('url', 'url_transparent'):
4577 return info
4578 if smuggled_data.get('is_music_url'):
4579 parsed_url = urllib.parse.urlparse(info['url'])
4580 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4581 smuggled_data.pop('is_music_url')
4582 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4583 if smuggled_data:
4584 info['url'] = smuggle_url(info['url'], smuggled_data)
4585 return info
4586
4587 @functools.wraps(func)
4588 def wrapper(self, url):
4589 url, smuggled_data = unsmuggle_url(url, {})
4590 if self.is_music_url(url):
4591 smuggled_data['is_music_url'] = True
4592 info_dict = func(self, url, smuggled_data)
4593 if smuggled_data:
4594 _smuggle(info_dict, smuggled_data)
4595 if info_dict.get('entries'):
4596 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4597 return info_dict
4598 return wrapper
4599
4600 @staticmethod
4601 def _extract_basic_item_renderer(item):
4602 # Modified from _extract_grid_item_renderer
4603 known_basic_renderers = (
4604 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4605 )
4606 for key, renderer in item.items():
4607 if not isinstance(renderer, dict):
4608 continue
4609 elif key in known_basic_renderers:
4610 return renderer
4611 elif key.startswith('grid') and key.endswith('Renderer'):
4612 return renderer
4613
4614 def _extract_channel_renderer(self, renderer):
4615 channel_id = self.ucid_or_none(renderer['channelId'])
4616 title = self._get_text(renderer, 'title')
4617 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4618 channel_handle = self.handle_from_url(
4619 traverse_obj(renderer, (
4620 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4621 ('browseEndpoint', 'canonicalBaseUrl')),
4622 {str}), get_all=False))
4623 if not channel_handle:
4624 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4625 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4626 return {
4627 '_type': 'url',
4628 'url': channel_url,
4629 'id': channel_id,
4630 'ie_key': YoutubeTabIE.ie_key(),
4631 'channel': title,
4632 'uploader': title,
4633 'channel_id': channel_id,
4634 'channel_url': channel_url,
4635 'title': title,
4636 'uploader_id': channel_handle,
4637 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4638 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4639 # However, in feed/channels this is set correctly to the subscriber count
4640 'channel_follower_count': traverse_obj(
4641 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4642 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4643 'playlist_count': (
4644 # videoCountText may be the subscriber count
4645 self._get_count(renderer, 'videoCountText')
4646 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4647 'description': self._get_text(renderer, 'descriptionSnippet'),
4648 'channel_is_verified': True if self._has_badge(
4649 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4650 }
4651
4652 def _grid_entries(self, grid_renderer):
4653 for item in grid_renderer['items']:
4654 if not isinstance(item, dict):
4655 continue
4656 renderer = self._extract_basic_item_renderer(item)
4657 if not isinstance(renderer, dict):
4658 continue
4659 title = self._get_text(renderer, 'title')
4660
4661 # playlist
4662 playlist_id = renderer.get('playlistId')
4663 if playlist_id:
4664 yield self.url_result(
4665 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4666 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4667 video_title=title)
4668 continue
4669 # video
4670 video_id = renderer.get('videoId')
4671 if video_id:
4672 yield self._extract_video(renderer)
4673 continue
4674 # channel
4675 channel_id = renderer.get('channelId')
4676 if channel_id:
4677 yield self._extract_channel_renderer(renderer)
4678 continue
4679 # generic endpoint URL support
4680 ep_url = urljoin('https://www.youtube.com/', try_get(
4681 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4682 str))
4683 if ep_url:
4684 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4685 if ie.suitable(ep_url):
4686 yield self.url_result(
4687 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4688 break
4689
4690 def _music_reponsive_list_entry(self, renderer):
4691 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4692 if video_id:
4693 title = traverse_obj(renderer, (
4694 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4695 'text', 'runs', 0, 'text'))
4696 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4697 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4698 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4699 if playlist_id:
4700 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4701 if video_id:
4702 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4703 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4704 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4705 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4706 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4707 if browse_id:
4708 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4709 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4710
4711 def _shelf_entries_from_content(self, shelf_renderer):
4712 content = shelf_renderer.get('content')
4713 if not isinstance(content, dict):
4714 return
4715 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4716 if renderer:
4717 # TODO: add support for nested playlists so each shelf is processed
4718 # as separate playlist
4719 # TODO: this includes only first N items
4720 yield from self._grid_entries(renderer)
4721 renderer = content.get('horizontalListRenderer')
4722 if renderer:
4723 # TODO
4724 pass
4725
4726 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4727 ep = try_get(
4728 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4729 str)
4730 shelf_url = urljoin('https://www.youtube.com', ep)
4731 if shelf_url:
4732 # Skipping links to another channels, note that checking for
4733 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4734 # will not work
4735 if skip_channels and '/channels?' in shelf_url:
4736 return
4737 title = self._get_text(shelf_renderer, 'title')
4738 yield self.url_result(shelf_url, video_title=title)
4739 # Shelf may not contain shelf URL, fallback to extraction from content
4740 yield from self._shelf_entries_from_content(shelf_renderer)
4741
4742 def _playlist_entries(self, video_list_renderer):
4743 for content in video_list_renderer['contents']:
4744 if not isinstance(content, dict):
4745 continue
4746 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4747 if not isinstance(renderer, dict):
4748 continue
4749 video_id = renderer.get('videoId')
4750 if not video_id:
4751 continue
4752 yield self._extract_video(renderer)
4753
4754 def _rich_entries(self, rich_grid_renderer):
4755 renderer = traverse_obj(
4756 rich_grid_renderer,
4757 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4758 video_id = renderer.get('videoId')
4759 if video_id:
4760 yield self._extract_video(renderer)
4761 return
4762 playlist_id = renderer.get('playlistId')
4763 if playlist_id:
4764 yield self.url_result(
4765 f'https://www.youtube.com/playlist?list={playlist_id}',
4766 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4767 video_title=self._get_text(renderer, 'title'))
4768 return
4769
4770 def _video_entry(self, video_renderer):
4771 video_id = video_renderer.get('videoId')
4772 if video_id:
4773 return self._extract_video(video_renderer)
4774
4775 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4776 url = urljoin('https://youtube.com', traverse_obj(
4777 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4778 if url:
4779 return self.url_result(
4780 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4781
4782 def _post_thread_entries(self, post_thread_renderer):
4783 post_renderer = try_get(
4784 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4785 if not post_renderer:
4786 return
4787 # video attachment
4788 video_renderer = try_get(
4789 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4790 video_id = video_renderer.get('videoId')
4791 if video_id:
4792 entry = self._extract_video(video_renderer)
4793 if entry:
4794 yield entry
4795 # playlist attachment
4796 playlist_id = try_get(
4797 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4798 if playlist_id:
4799 yield self.url_result(
4800 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4801 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4802 # inline video links
4803 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4804 for run in runs:
4805 if not isinstance(run, dict):
4806 continue
4807 ep_url = try_get(
4808 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4809 if not ep_url:
4810 continue
4811 if not YoutubeIE.suitable(ep_url):
4812 continue
4813 ep_video_id = YoutubeIE._match_id(ep_url)
4814 if video_id == ep_video_id:
4815 continue
4816 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4817
4818 def _post_thread_continuation_entries(self, post_thread_continuation):
4819 contents = post_thread_continuation.get('contents')
4820 if not isinstance(contents, list):
4821 return
4822 for content in contents:
4823 renderer = content.get('backstagePostThreadRenderer')
4824 if isinstance(renderer, dict):
4825 yield from self._post_thread_entries(renderer)
4826 continue
4827 renderer = content.get('videoRenderer')
4828 if isinstance(renderer, dict):
4829 yield self._video_entry(renderer)
4830
4831 r''' # unused
4832 def _rich_grid_entries(self, contents):
4833 for content in contents:
4834 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4835 if video_renderer:
4836 entry = self._video_entry(video_renderer)
4837 if entry:
4838 yield entry
4839 '''
4840
4841 def _report_history_entries(self, renderer):
4842 for url in traverse_obj(renderer, (
4843 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4844 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4845 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4846 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4847
4848 def _extract_entries(self, parent_renderer, continuation_list):
4849 # continuation_list is modified in-place with continuation_list = [continuation_token]
4850 continuation_list[:] = [None]
4851 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4852 for content in contents:
4853 if not isinstance(content, dict):
4854 continue
4855 is_renderer = traverse_obj(
4856 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4857 expected_type=dict)
4858 if not is_renderer:
4859 if content.get('richItemRenderer'):
4860 for entry in self._rich_entries(content['richItemRenderer']):
4861 yield entry
4862 continuation_list[0] = self._extract_continuation(parent_renderer)
4863 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4864 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4865 yield from self._report_history_entries(table)
4866 continuation_list[0] = self._extract_continuation(table)
4867 continue
4868
4869 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4870 for isr_content in isr_contents:
4871 if not isinstance(isr_content, dict):
4872 continue
4873
4874 known_renderers = {
4875 'playlistVideoListRenderer': self._playlist_entries,
4876 'gridRenderer': self._grid_entries,
4877 'reelShelfRenderer': self._grid_entries,
4878 'shelfRenderer': self._shelf_entries,
4879 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4880 'backstagePostThreadRenderer': self._post_thread_entries,
4881 'videoRenderer': lambda x: [self._video_entry(x)],
4882 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4883 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4884 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
4885 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
4886 }
4887 for key, renderer in isr_content.items():
4888 if key not in known_renderers:
4889 continue
4890 for entry in known_renderers[key](renderer):
4891 if entry:
4892 yield entry
4893 continuation_list[0] = self._extract_continuation(renderer)
4894 break
4895
4896 if not continuation_list[0]:
4897 continuation_list[0] = self._extract_continuation(is_renderer)
4898
4899 if not continuation_list[0]:
4900 continuation_list[0] = self._extract_continuation(parent_renderer)
4901
4902 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4903 continuation_list = [None]
4904 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4905 tab_content = try_get(tab, lambda x: x['content'], dict)
4906 if not tab_content:
4907 return
4908 parent_renderer = (
4909 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4910 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4911 yield from extract_entries(parent_renderer)
4912 continuation = continuation_list[0]
4913
4914 for page_num in itertools.count(1):
4915 if not continuation:
4916 break
4917 headers = self.generate_api_headers(
4918 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4919 response = self._extract_response(
4920 item_id=f'{item_id} page {page_num}',
4921 query=continuation, headers=headers, ytcfg=ytcfg,
4922 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4923
4924 if not response:
4925 break
4926 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4927 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4928 visitor_data = self._extract_visitor_data(response) or visitor_data
4929
4930 known_renderers = {
4931 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4932 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4933 'gridVideoRenderer': (self._grid_entries, 'items'),
4934 'gridChannelRenderer': (self._grid_entries, 'items'),
4935 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4936 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4937 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4938 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4939 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4940 'playlistVideoListContinuation': (self._playlist_entries, None),
4941 'gridContinuation': (self._grid_entries, None),
4942 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4943 'sectionListContinuation': (extract_entries, None), # for feeds
4944 }
4945
4946 continuation_items = traverse_obj(response, (
4947 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4948 'appendContinuationItemsAction', 'continuationItems'
4949 ), 'continuationContents', get_all=False)
4950 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4951
4952 video_items_renderer = None
4953 for key in continuation_item.keys():
4954 if key not in known_renderers:
4955 continue
4956 func, parent_key = known_renderers[key]
4957 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4958 continuation_list = [None]
4959 yield from func(video_items_renderer)
4960 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4961
4962 if not video_items_renderer:
4963 break
4964
4965 @staticmethod
4966 def _extract_selected_tab(tabs, fatal=True):
4967 for tab_renderer in tabs:
4968 if tab_renderer.get('selected'):
4969 return tab_renderer
4970 if fatal:
4971 raise ExtractorError('Unable to find selected tab')
4972
4973 @staticmethod
4974 def _extract_tab_renderers(response):
4975 return traverse_obj(
4976 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4977
4978 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4979 metadata = self._extract_metadata_from_tabs(item_id, data)
4980
4981 selected_tab = self._extract_selected_tab(tabs)
4982 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4983 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
4984
4985 return self.playlist_result(
4986 self._entries(
4987 selected_tab, metadata['id'], ytcfg,
4988 self._extract_account_syncid(ytcfg, data),
4989 self._extract_visitor_data(data, ytcfg)),
4990 **metadata)
4991
4992 def _extract_metadata_from_tabs(self, item_id, data):
4993 info = {'id': item_id}
4994
4995 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
4996 if metadata_renderer:
4997 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
4998 ('channelUrl', {self.ucid_from_url}))
4999 info.update({
5000 'channel': metadata_renderer.get('title'),
5001 'channel_id': channel_id,
5002 })
5003 if info['channel_id']:
5004 info['id'] = info['channel_id']
5005 else:
5006 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5007
5008 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5009 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5010 def _get_uncropped(url):
5011 return url_or_none((url or '').split('=')[0] + '=s0')
5012
5013 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5014 if avatar_thumbnails:
5015 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5016 if uncropped_avatar:
5017 avatar_thumbnails.append({
5018 'url': uncropped_avatar,
5019 'id': 'avatar_uncropped',
5020 'preference': 1
5021 })
5022
5023 channel_banners = self._extract_thumbnails(
5024 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5025 for banner in channel_banners:
5026 banner['preference'] = -10
5027
5028 if channel_banners:
5029 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5030 if uncropped_banner:
5031 channel_banners.append({
5032 'url': uncropped_banner,
5033 'id': 'banner_uncropped',
5034 'preference': -5
5035 })
5036
5037 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5038 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5039 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5040
5041 primary_thumbnails = self._extract_thumbnails(
5042 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5043 playlist_thumbnails = self._extract_thumbnails(
5044 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5045
5046 info.update({
5047 'title': (traverse_obj(metadata_renderer, 'title')
5048 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5049 or info['id']),
5050 'availability': self._extract_availability(data),
5051 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
5052 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5053 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
5054 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5055 })
5056
5057 channel_handle = (
5058 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5059 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5060
5061 if channel_handle:
5062 info.update({
5063 'uploader_id': channel_handle,
5064 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5065 })
5066
5067 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5068 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5069 info['channel_is_verified'] = True
5070 # Playlist stats is a text runs array containing [video count, view count, last updated].
5071 # last updated or (view count and last updated) may be missing.
5072 playlist_stats = get_first(
5073 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5074
5075 last_updated_unix = self._parse_time_text(
5076 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5077 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5078 info['modified_date'] = strftime_or_none(last_updated_unix)
5079
5080 info['view_count'] = self._get_count(playlist_stats, 1)
5081 if info['view_count'] is None: # 0 is allowed
5082 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5083 if info['view_count'] is None:
5084 info['view_count'] = self._get_count(data, (
5085 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5086 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5087
5088 info['playlist_count'] = self._get_count(playlist_stats, 0)
5089 if info['playlist_count'] is None: # 0 is allowed
5090 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5091
5092 if not info.get('channel_id'):
5093 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5094 if not owner: # Deprecated
5095 owner = traverse_obj(
5096 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5097 ('videoOwner', 'videoOwnerRenderer', 'title'))
5098 owner_text = self._get_text(owner)
5099 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5100 info.update({
5101 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5102 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5103 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5104 })
5105
5106 info.update({
5107 'uploader': info['channel'],
5108 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5109 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5110 })
5111
5112 return info
5113
5114 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5115 first_id = last_id = response = None
5116 for page_num in itertools.count(1):
5117 videos = list(self._playlist_entries(playlist))
5118 if not videos:
5119 return
5120 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5121 if start >= len(videos):
5122 return
5123 yield from videos[start:]
5124 first_id = first_id or videos[0]['id']
5125 last_id = videos[-1]['id']
5126 watch_endpoint = try_get(
5127 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5128 headers = self.generate_api_headers(
5129 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5130 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5131 query = {
5132 'playlistId': playlist_id,
5133 'videoId': watch_endpoint.get('videoId') or last_id,
5134 'index': watch_endpoint.get('index') or len(videos),
5135 'params': watch_endpoint.get('params') or 'OAE%3D'
5136 }
5137 response = self._extract_response(
5138 item_id='%s page %d' % (playlist_id, page_num),
5139 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5140 check_get_keys='contents'
5141 )
5142 playlist = try_get(
5143 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5144
5145 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5146 title = playlist.get('title') or try_get(
5147 data, lambda x: x['titleText']['simpleText'], str)
5148 playlist_id = playlist.get('playlistId') or item_id
5149
5150 # Delegating everything except mix playlists to regular tab-based playlist URL
5151 playlist_url = urljoin(url, try_get(
5152 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5153 str))
5154
5155 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5156 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5157 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5158
5159 if playlist_url and playlist_url != url and not is_known_unviewable:
5160 return self.url_result(
5161 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5162 video_title=title)
5163
5164 return self.playlist_result(
5165 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5166 playlist_id=playlist_id, playlist_title=title)
5167
5168 def _extract_availability(self, data):
5169 """
5170 Gets the availability of a given playlist/tab.
5171 Note: Unless YouTube tells us explicitly, we do not assume it is public
5172 @param data: response
5173 """
5174 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5175 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5176 player_header_privacy = playlist_header_renderer.get('privacy')
5177
5178 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5179
5180 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5181 privacy_setting_icon = get_first(
5182 (playlist_header_renderer, sidebar_renderer),
5183 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5184 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5185 expected_type=str)
5186
5187 microformats_is_unlisted = traverse_obj(
5188 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5189
5190 return (
5191 'public' if (
5192 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5193 or player_header_privacy == 'PUBLIC'
5194 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5195 else self._availability(
5196 is_private=(
5197 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5198 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5199 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5200 is_unlisted=(
5201 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5202 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5203 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5204 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5205 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5206 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5207 needs_auth=False))
5208
5209 @staticmethod
5210 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5211 sidebar_renderer = try_get(
5212 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5213 for item in sidebar_renderer:
5214 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5215 if renderer:
5216 return renderer
5217
5218 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5219 """
5220 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5221 """
5222 is_playlist = bool(traverse_obj(
5223 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5224 if not is_playlist:
5225 return
5226 headers = self.generate_api_headers(
5227 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5228 visitor_data=self._extract_visitor_data(data, ytcfg))
5229 query = {
5230 'params': 'wgYCCAA=',
5231 'browseId': f'VL{item_id}'
5232 }
5233 return self._extract_response(
5234 item_id=item_id, headers=headers, query=query,
5235 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5236 note='Redownloading playlist API JSON with unavailable videos')
5237
5238 @functools.cached_property
5239 def skip_webpage(self):
5240 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5241
5242 def _extract_webpage(self, url, item_id, fatal=True):
5243 webpage, data = None, None
5244 for retry in self.RetryManager(fatal=fatal):
5245 try:
5246 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5247 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5248 except ExtractorError as e:
5249 if isinstance(e.cause, network_exceptions):
5250 if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
5251 retry.error = e
5252 continue
5253 self._error_or_warning(e, fatal=fatal)
5254 break
5255
5256 try:
5257 self._extract_and_report_alerts(data)
5258 except ExtractorError as e:
5259 self._error_or_warning(e, fatal=fatal)
5260 break
5261
5262 # Sometimes youtube returns a webpage with incomplete ytInitialData
5263 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5264 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5265 retry.error = ExtractorError('Incomplete yt initial data received')
5266 continue
5267
5268 return webpage, data
5269
5270 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5271 """Use if failed to extract ytcfg (and data) from initial webpage"""
5272 if not ytcfg and self.is_authenticated:
5273 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5274 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5275 raise ExtractorError(
5276 f'{msg}. If you are not downloading private content, or '
5277 'your cookies are only for the first account and channel,'
5278 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5279 expected=True)
5280 self.report_warning(msg, only_once=True)
5281
5282 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5283 data = None
5284 if not self.skip_webpage:
5285 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5286 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5287 # Reject webpage data if redirected to home page without explicitly requesting
5288 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5289 if (url != 'https://www.youtube.com/feed/recommended'
5290 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5291 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5292 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5293 if fatal:
5294 raise ExtractorError(msg, expected=True)
5295 self.report_warning(msg, only_once=True)
5296 if not data:
5297 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5298 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5299 return data, ytcfg
5300
5301 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5302 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5303 resolve_response = self._extract_response(
5304 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5305 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5306 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5307 for ep_key, ep in endpoints.items():
5308 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5309 if params:
5310 return self._extract_response(
5311 item_id=item_id, query=params, ep=ep, headers=headers,
5312 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5313 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5314 err_note = 'Failed to resolve url (does the playlist exist?)'
5315 if fatal:
5316 raise ExtractorError(err_note, expected=True)
5317 self.report_warning(err_note, item_id)
5318
5319 _SEARCH_PARAMS = None
5320
5321 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5322 data = {'query': query}
5323 if params is NO_DEFAULT:
5324 params = self._SEARCH_PARAMS
5325 if params:
5326 data['params'] = params
5327
5328 content_keys = (
5329 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5330 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5331 # ytmusic search
5332 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5333 ('continuationContents', ),
5334 )
5335 display_id = f'query "{query}"'
5336 check_get_keys = tuple({keys[0] for keys in content_keys})
5337 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5338 self._report_playlist_authcheck(ytcfg, fatal=False)
5339
5340 continuation_list = [None]
5341 search = None
5342 for page_num in itertools.count(1):
5343 data.update(continuation_list[0] or {})
5344 headers = self.generate_api_headers(
5345 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5346 search = self._extract_response(
5347 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5348 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5349 slr_contents = traverse_obj(search, *content_keys)
5350 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5351 if not continuation_list[0]:
5352 break
5353
5354
5355 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5356 IE_DESC = 'YouTube Tabs'
5357 _VALID_URL = r'''(?x:
5358 https?://
5359 (?!consent\.)(?:\w+\.)?
5360 (?:
5361 youtube(?:kids)?\.com|
5362 %(invidious)s
5363 )/
5364 (?:
5365 (?P<channel_type>channel|c|user|browse)/|
5366 (?P<not_channel>
5367 feed/|hashtag/|
5368 (?:playlist|watch)\?.*?\blist=
5369 )|
5370 (?!(?:%(reserved_names)s)\b) # Direct URLs
5371 )
5372 (?P<id>[^/?\#&]+)
5373 )''' % {
5374 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5375 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5376 }
5377 IE_NAME = 'youtube:tab'
5378
5379 _TESTS = [{
5380 'note': 'playlists, multipage',
5381 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5382 'playlist_mincount': 94,
5383 'info_dict': {
5384 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5385 'title': 'Igor Kleiner - Playlists',
5386 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5387 'uploader': 'Igor Kleiner',
5388 'uploader_id': '@IgorDataScience',
5389 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5390 'channel': 'Igor Kleiner',
5391 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5392 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5393 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5394 'channel_follower_count': int
5395 },
5396 }, {
5397 'note': 'playlists, multipage, different order',
5398 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5399 'playlist_mincount': 94,
5400 'info_dict': {
5401 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5402 'title': 'Igor Kleiner - Playlists',
5403 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5404 'uploader': 'Igor Kleiner',
5405 'uploader_id': '@IgorDataScience',
5406 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5407 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5408 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5409 'channel': 'Igor Kleiner',
5410 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5411 'channel_follower_count': int
5412 },
5413 }, {
5414 'note': 'playlists, series',
5415 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5416 'playlist_mincount': 5,
5417 'info_dict': {
5418 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5419 'title': '3Blue1Brown - Playlists',
5420 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5421 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5422 'channel': '3Blue1Brown',
5423 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5424 'uploader_id': '@3blue1brown',
5425 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5426 'uploader': '3Blue1Brown',
5427 'tags': ['Mathematics'],
5428 'channel_follower_count': int,
5429 'channel_is_verified': True,
5430 },
5431 }, {
5432 'note': 'playlists, singlepage',
5433 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5434 'playlist_mincount': 4,
5435 'info_dict': {
5436 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5437 'title': 'ThirstForScience - Playlists',
5438 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5439 'uploader': 'ThirstForScience',
5440 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5441 'uploader_id': '@ThirstForScience',
5442 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5443 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5444 'tags': 'count:13',
5445 'channel': 'ThirstForScience',
5446 'channel_follower_count': int
5447 }
5448 }, {
5449 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5450 'only_matching': True,
5451 }, {
5452 'note': 'basic, single video playlist',
5453 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5454 'info_dict': {
5455 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5456 'title': 'youtube-dl public playlist',
5457 'description': '',
5458 'tags': [],
5459 'view_count': int,
5460 'modified_date': '20201130',
5461 'channel': 'Sergey M.',
5462 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5463 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5464 'availability': 'public',
5465 'uploader': 'Sergey M.',
5466 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5467 'uploader_id': '@sergeym.6173',
5468 },
5469 'playlist_count': 1,
5470 }, {
5471 'note': 'empty playlist',
5472 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5473 'info_dict': {
5474 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5475 'title': 'youtube-dl empty playlist',
5476 'tags': [],
5477 'channel': 'Sergey M.',
5478 'description': '',
5479 'modified_date': '20160902',
5480 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5481 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5482 'availability': 'public',
5483 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5484 'uploader_id': '@sergeym.6173',
5485 'uploader': 'Sergey M.',
5486 },
5487 'playlist_count': 0,
5488 }, {
5489 'note': 'Home tab',
5490 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5491 'info_dict': {
5492 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5493 'title': 'lex will - Home',
5494 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5495 'uploader': 'lex will',
5496 'uploader_id': '@lexwill718',
5497 'channel': 'lex will',
5498 'tags': ['bible', 'history', 'prophesy'],
5499 'uploader_url': 'https://www.youtube.com/@lexwill718',
5500 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5501 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5502 'channel_follower_count': int
5503 },
5504 'playlist_mincount': 2,
5505 }, {
5506 'note': 'Videos tab',
5507 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5508 'info_dict': {
5509 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5510 'title': 'lex will - Videos',
5511 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5512 'uploader': 'lex will',
5513 'uploader_id': '@lexwill718',
5514 'tags': ['bible', 'history', 'prophesy'],
5515 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5516 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5517 'uploader_url': 'https://www.youtube.com/@lexwill718',
5518 'channel': 'lex will',
5519 'channel_follower_count': int
5520 },
5521 'playlist_mincount': 975,
5522 }, {
5523 'note': 'Videos tab, sorted by popular',
5524 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5525 'info_dict': {
5526 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5527 'title': 'lex will - Videos',
5528 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5529 'uploader': 'lex will',
5530 'uploader_id': '@lexwill718',
5531 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5532 'uploader_url': 'https://www.youtube.com/@lexwill718',
5533 'channel': 'lex will',
5534 'tags': ['bible', 'history', 'prophesy'],
5535 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5536 'channel_follower_count': int
5537 },
5538 'playlist_mincount': 199,
5539 }, {
5540 'note': 'Playlists tab',
5541 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5542 'info_dict': {
5543 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5544 'title': 'lex will - Playlists',
5545 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5546 'uploader': 'lex will',
5547 'uploader_id': '@lexwill718',
5548 'uploader_url': 'https://www.youtube.com/@lexwill718',
5549 'channel': 'lex will',
5550 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5551 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5552 'tags': ['bible', 'history', 'prophesy'],
5553 'channel_follower_count': int
5554 },
5555 'playlist_mincount': 17,
5556 }, {
5557 'note': 'Community tab',
5558 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5559 'info_dict': {
5560 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5561 'title': 'lex will - Community',
5562 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5563 'channel': 'lex will',
5564 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5565 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5566 'tags': ['bible', 'history', 'prophesy'],
5567 'channel_follower_count': int,
5568 'uploader_url': 'https://www.youtube.com/@lexwill718',
5569 'uploader_id': '@lexwill718',
5570 'uploader': 'lex will',
5571 },
5572 'playlist_mincount': 18,
5573 }, {
5574 'note': 'Channels tab',
5575 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5576 'info_dict': {
5577 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5578 'title': 'lex will - Channels',
5579 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5580 'channel': 'lex will',
5581 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5582 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5583 'tags': ['bible', 'history', 'prophesy'],
5584 'channel_follower_count': int,
5585 'uploader_url': 'https://www.youtube.com/@lexwill718',
5586 'uploader_id': '@lexwill718',
5587 'uploader': 'lex will',
5588 },
5589 'playlist_mincount': 12,
5590 }, {
5591 'note': 'Search tab',
5592 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5593 'playlist_mincount': 40,
5594 'info_dict': {
5595 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5596 'title': '3Blue1Brown - Search - linear algebra',
5597 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5598 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5599 'tags': ['Mathematics'],
5600 'channel': '3Blue1Brown',
5601 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5602 'channel_follower_count': int,
5603 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5604 'uploader_id': '@3blue1brown',
5605 'uploader': '3Blue1Brown',
5606 'channel_is_verified': True,
5607 },
5608 }, {
5609 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5610 'only_matching': True,
5611 }, {
5612 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5613 'only_matching': True,
5614 }, {
5615 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5616 'only_matching': True,
5617 }, {
5618 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5619 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5620 'info_dict': {
5621 'title': '29C3: Not my department',
5622 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5623 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5624 'tags': [],
5625 'view_count': int,
5626 'modified_date': '20150605',
5627 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5628 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5629 'channel': 'Christiaan008',
5630 'availability': 'public',
5631 'uploader_id': '@ChRiStIaAn008',
5632 'uploader': 'Christiaan008',
5633 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5634 },
5635 'playlist_count': 96,
5636 }, {
5637 'note': 'Large playlist',
5638 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5639 'info_dict': {
5640 'title': 'Uploads from Cauchemar',
5641 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5642 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5643 'tags': [],
5644 'modified_date': r're:\d{8}',
5645 'channel': 'Cauchemar',
5646 'view_count': int,
5647 'description': '',
5648 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5649 'availability': 'public',
5650 'uploader_id': '@Cauchemar89',
5651 'uploader': 'Cauchemar',
5652 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5653 },
5654 'playlist_mincount': 1123,
5655 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5656 }, {
5657 'note': 'even larger playlist, 8832 videos',
5658 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5659 'only_matching': True,
5660 }, {
5661 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5662 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5663 'info_dict': {
5664 'title': 'Uploads from Interstellar Movie',
5665 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5666 'tags': [],
5667 'view_count': int,
5668 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5669 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5670 'channel': 'Interstellar Movie',
5671 'description': '',
5672 'modified_date': r're:\d{8}',
5673 'availability': 'public',
5674 'uploader_id': '@InterstellarMovie',
5675 'uploader': 'Interstellar Movie',
5676 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5677 },
5678 'playlist_mincount': 21,
5679 }, {
5680 'note': 'Playlist with "show unavailable videos" button',
5681 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5682 'info_dict': {
5683 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5684 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5685 'view_count': int,
5686 'channel': 'Phim Siêu Nhân Nhật Bản',
5687 'tags': [],
5688 'description': '',
5689 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5690 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5691 'modified_date': r're:\d{8}',
5692 'availability': 'public',
5693 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5694 'uploader_id': '@phimsieunhannhatban',
5695 'uploader': 'Phim Siêu Nhân Nhật Bản',
5696 },
5697 'playlist_mincount': 200,
5698 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5699 }, {
5700 'note': 'Playlist with unavailable videos in page 7',
5701 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5702 'info_dict': {
5703 'title': 'Uploads from BlankTV',
5704 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5705 'channel': 'BlankTV',
5706 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5707 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5708 'view_count': int,
5709 'tags': [],
5710 'modified_date': r're:\d{8}',
5711 'description': '',
5712 'availability': 'public',
5713 'uploader_id': '@blanktv',
5714 'uploader': 'BlankTV',
5715 'uploader_url': 'https://www.youtube.com/@blanktv',
5716 },
5717 'playlist_mincount': 1000,
5718 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5719 }, {
5720 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5721 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5722 'info_dict': {
5723 'title': 'Data Analysis with Dr Mike Pound',
5724 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5725 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5726 'tags': [],
5727 'view_count': int,
5728 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5729 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5730 'channel': 'Computerphile',
5731 'availability': 'public',
5732 'modified_date': '20190712',
5733 'uploader_id': '@Computerphile',
5734 'uploader': 'Computerphile',
5735 'uploader_url': 'https://www.youtube.com/@Computerphile',
5736 },
5737 'playlist_mincount': 11,
5738 }, {
5739 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5740 'only_matching': True,
5741 }, {
5742 'note': 'Playlist URL that does not actually serve a playlist',
5743 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5744 'info_dict': {
5745 'id': 'FqZTN594JQw',
5746 'ext': 'webm',
5747 'title': "Smiley's People 01 detective, Adventure Series, Action",
5748 'upload_date': '20150526',
5749 'license': 'Standard YouTube License',
5750 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5751 'categories': ['People & Blogs'],
5752 'tags': list,
5753 'view_count': int,
5754 'like_count': int,
5755 },
5756 'params': {
5757 'skip_download': True,
5758 },
5759 'skip': 'This video is not available.',
5760 'add_ie': [YoutubeIE.ie_key()],
5761 }, {
5762 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5763 'only_matching': True,
5764 }, {
5765 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5766 'only_matching': True,
5767 }, {
5768 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5769 'info_dict': {
5770 'id': 'hGkQjiJLjWQ', # This will keep changing
5771 'ext': 'mp4',
5772 'title': str,
5773 'upload_date': r're:\d{8}',
5774 'description': str,
5775 'categories': ['News & Politics'],
5776 'tags': list,
5777 'like_count': int,
5778 'release_timestamp': int,
5779 'channel': 'Sky News',
5780 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5781 'age_limit': 0,
5782 'view_count': int,
5783 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5784 'playable_in_embed': True,
5785 'release_date': r're:\d+',
5786 'availability': 'public',
5787 'live_status': 'is_live',
5788 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5789 'channel_follower_count': int,
5790 'concurrent_view_count': int,
5791 'uploader_url': 'https://www.youtube.com/@SkyNews',
5792 'uploader_id': '@SkyNews',
5793 'uploader': 'Sky News',
5794 'channel_is_verified': True,
5795 },
5796 'params': {
5797 'skip_download': True,
5798 },
5799 'expected_warnings': ['Ignoring subtitle tracks found in '],
5800 }, {
5801 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5802 'info_dict': {
5803 'id': 'a48o2S1cPoo',
5804 'ext': 'mp4',
5805 'title': 'The Young Turks - Live Main Show',
5806 'upload_date': '20150715',
5807 'license': 'Standard YouTube License',
5808 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5809 'categories': ['News & Politics'],
5810 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5811 'like_count': int,
5812 },
5813 'params': {
5814 'skip_download': True,
5815 },
5816 'only_matching': True,
5817 }, {
5818 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5819 'only_matching': True,
5820 }, {
5821 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5822 'only_matching': True,
5823 }, {
5824 'note': 'A channel that is not live. Should raise error',
5825 'url': 'https://www.youtube.com/user/numberphile/live',
5826 'only_matching': True,
5827 }, {
5828 'url': 'https://www.youtube.com/feed/trending',
5829 'only_matching': True,
5830 }, {
5831 'url': 'https://www.youtube.com/feed/library',
5832 'only_matching': True,
5833 }, {
5834 'url': 'https://www.youtube.com/feed/history',
5835 'only_matching': True,
5836 }, {
5837 'url': 'https://www.youtube.com/feed/subscriptions',
5838 'only_matching': True,
5839 }, {
5840 'url': 'https://www.youtube.com/feed/watch_later',
5841 'only_matching': True,
5842 }, {
5843 'note': 'Recommended - redirects to home page.',
5844 'url': 'https://www.youtube.com/feed/recommended',
5845 'only_matching': True,
5846 }, {
5847 'note': 'inline playlist with not always working continuations',
5848 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5849 'only_matching': True,
5850 }, {
5851 'url': 'https://www.youtube.com/course',
5852 'only_matching': True,
5853 }, {
5854 'url': 'https://www.youtube.com/zsecurity',
5855 'only_matching': True,
5856 }, {
5857 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5858 'only_matching': True,
5859 }, {
5860 'url': 'https://www.youtube.com/TheYoungTurks/live',
5861 'only_matching': True,
5862 }, {
5863 'url': 'https://www.youtube.com/hashtag/cctv9',
5864 'info_dict': {
5865 'id': 'cctv9',
5866 'title': '#cctv9',
5867 'tags': [],
5868 },
5869 'playlist_mincount': 300, # not consistent but should be over 300
5870 }, {
5871 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5872 'only_matching': True,
5873 }, {
5874 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5875 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5876 'only_matching': True
5877 }, {
5878 'note': '/browse/ should redirect to /channel/',
5879 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5880 'only_matching': True
5881 }, {
5882 'note': 'VLPL, should redirect to playlist?list=PL...',
5883 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5884 'info_dict': {
5885 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5886 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5887 'title': 'NCS : All Releases 💿',
5888 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5889 'modified_date': r're:\d{8}',
5890 'view_count': int,
5891 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5892 'tags': [],
5893 'channel': 'NoCopyrightSounds',
5894 'availability': 'public',
5895 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5896 'uploader': 'NoCopyrightSounds',
5897 'uploader_id': '@NoCopyrightSounds',
5898 },
5899 'playlist_mincount': 166,
5900 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5901 }, {
5902 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5903 'note': 'Topic, should redirect to playlist?list=UU...',
5904 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5905 'info_dict': {
5906 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5907 'title': 'Uploads from Royalty Free Music - Topic',
5908 'tags': [],
5909 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5910 'channel': 'Royalty Free Music - Topic',
5911 'view_count': int,
5912 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5913 'modified_date': r're:\d{8}',
5914 'description': '',
5915 'availability': 'public',
5916 'uploader': 'Royalty Free Music - Topic',
5917 },
5918 'playlist_mincount': 101,
5919 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5920 }, {
5921 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5922 # Treat as a general feed
5923 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5924 'info_dict': {
5925 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5926 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5927 'tags': [],
5928 },
5929 'playlist_mincount': 9,
5930 }, {
5931 'note': 'Youtube music Album',
5932 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5933 'info_dict': {
5934 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5935 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5936 'tags': [],
5937 'view_count': int,
5938 'description': '',
5939 'availability': 'unlisted',
5940 'modified_date': r're:\d{8}',
5941 },
5942 'playlist_count': 50,
5943 'expected_warnings': ['YouTube Music is not directly supported'],
5944 }, {
5945 'note': 'unlisted single video playlist',
5946 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5947 'info_dict': {
5948 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5949 'title': 'yt-dlp unlisted playlist test',
5950 'availability': 'unlisted',
5951 'tags': [],
5952 'modified_date': '20220418',
5953 'channel': 'colethedj',
5954 'view_count': int,
5955 'description': '',
5956 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5957 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5958 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5959 'uploader_id': '@colethedj1894',
5960 'uploader': 'colethedj',
5961 },
5962 'playlist': [{
5963 'info_dict': {
5964 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
5965 'id': 'BaW_jenozKc',
5966 '_type': 'url',
5967 'ie_key': 'Youtube',
5968 'duration': 10,
5969 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
5970 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
5971 'view_count': int,
5972 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
5973 'channel': 'Philipp Hagemeister',
5974 'uploader_id': '@PhilippHagemeister',
5975 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
5976 'uploader': 'Philipp Hagemeister',
5977 }
5978 }],
5979 'playlist_count': 1,
5980 'params': {'extract_flat': True},
5981 }, {
5982 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5983 'url': 'https://www.youtube.com/feed/recommended',
5984 'info_dict': {
5985 'id': 'recommended',
5986 'title': 'recommended',
5987 'tags': [],
5988 },
5989 'playlist_mincount': 50,
5990 'params': {
5991 'skip_download': True,
5992 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5993 },
5994 }, {
5995 'note': 'API Fallback: /videos tab, sorted by oldest first',
5996 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5997 'info_dict': {
5998 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5999 'title': 'Cody\'sLab - Videos',
6000 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6001 'channel': 'Cody\'sLab',
6002 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6003 'tags': [],
6004 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6005 'channel_follower_count': int
6006 },
6007 'playlist_mincount': 650,
6008 'params': {
6009 'skip_download': True,
6010 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6011 },
6012 'skip': 'Query for sorting no longer works',
6013 }, {
6014 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6015 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6016 'info_dict': {
6017 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6018 'title': 'Uploads from Royalty Free Music - Topic',
6019 'modified_date': r're:\d{8}',
6020 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6021 'description': '',
6022 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6023 'tags': [],
6024 'channel': 'Royalty Free Music - Topic',
6025 'view_count': int,
6026 'availability': 'public',
6027 'uploader': 'Royalty Free Music - Topic',
6028 },
6029 'playlist_mincount': 101,
6030 'params': {
6031 'skip_download': True,
6032 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6033 },
6034 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6035 }, {
6036 'note': 'non-standard redirect to regional channel',
6037 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6038 'only_matching': True
6039 }, {
6040 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6041 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6042 'info_dict': {
6043 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6044 'modified_date': '20220407',
6045 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6046 'tags': [],
6047 'availability': 'unlisted',
6048 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6049 'channel': 'pukkandan',
6050 'description': 'Test for collaborative playlist',
6051 'title': 'yt-dlp test - collaborative playlist',
6052 'view_count': int,
6053 'uploader_url': 'https://www.youtube.com/@pukkandan',
6054 'uploader_id': '@pukkandan',
6055 'uploader': 'pukkandan',
6056 },
6057 'playlist_mincount': 2
6058 }, {
6059 'note': 'translated tab name',
6060 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6061 'info_dict': {
6062 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6063 'tags': [],
6064 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6065 'description': 'test description',
6066 'title': 'cole-dlp-test-acc - 再生リスト',
6067 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6068 'channel': 'cole-dlp-test-acc',
6069 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6070 'uploader_id': '@coletdjnz',
6071 'uploader': 'cole-dlp-test-acc',
6072 },
6073 'playlist_mincount': 1,
6074 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6075 'expected_warnings': ['Preferring "ja"'],
6076 }, {
6077 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6078 'note': 'preferred lang set with playlist with translated video titles',
6079 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6080 'info_dict': {
6081 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6082 'tags': [],
6083 'view_count': int,
6084 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6085 'channel': 'cole-dlp-test-acc',
6086 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6087 'description': 'test',
6088 'title': 'dlp test playlist',
6089 'availability': 'public',
6090 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6091 'uploader_id': '@coletdjnz',
6092 'uploader': 'cole-dlp-test-acc',
6093 },
6094 'playlist_mincount': 1,
6095 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6096 'expected_warnings': ['Preferring "ja"'],
6097 }, {
6098 # shorts audio pivot for 2GtVksBMYFM.
6099 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6100 'info_dict': {
6101 'id': 'sfv_audio_pivot',
6102 'title': 'sfv_audio_pivot',
6103 'tags': [],
6104 },
6105 'playlist_mincount': 50,
6106
6107 }, {
6108 # Channel with a real live tab (not to be mistaken with streams tab)
6109 # Do not treat like it should redirect to live stream
6110 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6111 'info_dict': {
6112 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6113 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6114 'tags': [],
6115 },
6116 'playlist_mincount': 20,
6117 }, {
6118 # Tab name is not the same as tab id
6119 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6120 'info_dict': {
6121 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6122 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6123 'tags': [],
6124 },
6125 'playlist_mincount': 8,
6126 }, {
6127 # Home tab id is literally home. Not to get mistaken with featured
6128 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6129 'info_dict': {
6130 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6131 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6132 'tags': [],
6133 },
6134 'playlist_mincount': 8,
6135 }, {
6136 # Should get three playlists for videos, shorts and streams tabs
6137 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6138 'info_dict': {
6139 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6140 'title': 'Polka Ch. 尾丸ポルカ',
6141 'channel_follower_count': int,
6142 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6143 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6144 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6145 'channel': 'Polka Ch. 尾丸ポルカ',
6146 'tags': 'count:35',
6147 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6148 'uploader': 'Polka Ch. 尾丸ポルカ',
6149 'uploader_id': '@OmaruPolka',
6150 },
6151 'playlist_count': 3,
6152 }, {
6153 # Shorts tab with channel with handle
6154 # TODO: fix channel description
6155 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6156 'info_dict': {
6157 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6158 'title': 'Not Just Bikes - Shorts',
6159 'tags': 'count:12',
6160 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6161 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6162 'channel_follower_count': int,
6163 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6164 'channel': 'Not Just Bikes',
6165 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6166 'uploader': 'Not Just Bikes',
6167 'uploader_id': '@NotJustBikes',
6168 },
6169 'playlist_mincount': 10,
6170 }, {
6171 # Streams tab
6172 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6173 'info_dict': {
6174 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6175 'title': '中村悠一 - Live',
6176 'tags': 'count:7',
6177 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6178 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6179 'channel': '中村悠一',
6180 'channel_follower_count': int,
6181 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6182 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6183 'uploader_id': '@Yuichi-Nakamura',
6184 'uploader': '中村悠一',
6185 },
6186 'playlist_mincount': 60,
6187 }, {
6188 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6189 # See test_youtube_lists
6190 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6191 'only_matching': True,
6192 }, {
6193 # No uploads and no UCID given. Should fail with no uploads error
6194 # See test_youtube_lists
6195 'url': 'https://www.youtube.com/news',
6196 'only_matching': True
6197 }, {
6198 # No videos tab but has a shorts tab
6199 'url': 'https://www.youtube.com/c/TKFShorts',
6200 'info_dict': {
6201 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6202 'title': 'Shorts Break - Shorts',
6203 'tags': 'count:48',
6204 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6205 'channel': 'Shorts Break',
6206 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6207 'channel_follower_count': int,
6208 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6209 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6210 'uploader': 'Shorts Break',
6211 'uploader_id': '@ShortsBreak_Official',
6212 },
6213 'playlist_mincount': 30,
6214 }, {
6215 # Trending Now Tab. tab id is empty
6216 'url': 'https://www.youtube.com/feed/trending',
6217 'info_dict': {
6218 'id': 'trending',
6219 'title': 'trending - Now',
6220 'tags': [],
6221 },
6222 'playlist_mincount': 30,
6223 }, {
6224 # Trending Gaming Tab. tab id is empty
6225 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6226 'info_dict': {
6227 'id': 'trending',
6228 'title': 'trending - Gaming',
6229 'tags': [],
6230 },
6231 'playlist_mincount': 30,
6232 }, {
6233 # Shorts url result in shorts tab
6234 # TODO: Fix channel id extraction
6235 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6236 'info_dict': {
6237 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6238 'title': 'cole-dlp-test-acc - Shorts',
6239 'channel': 'cole-dlp-test-acc',
6240 'description': 'test description',
6241 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6242 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6243 'tags': [],
6244 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6245 'uploader_id': '@coletdjnz',
6246 'uploader': 'cole-dlp-test-acc',
6247 },
6248 'playlist': [{
6249 'info_dict': {
6250 # Channel data is not currently available for short renderers (as of 2023-03-01)
6251 '_type': 'url',
6252 'ie_key': 'Youtube',
6253 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6254 'id': 'sSM9J5YH_60',
6255 'title': 'SHORT short',
6256 'view_count': int,
6257 'thumbnails': list,
6258 }
6259 }],
6260 'params': {'extract_flat': True},
6261 }, {
6262 # Live video status should be extracted
6263 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6264 'info_dict': {
6265 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6266 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6267 'tags': []
6268 },
6269 'playlist': [{
6270 'info_dict': {
6271 '_type': 'url',
6272 'ie_key': 'Youtube',
6273 'url': 'startswith:https://www.youtube.com/watch?v=',
6274 'id': str,
6275 'title': str,
6276 'live_status': 'is_live',
6277 'channel_id': str,
6278 'channel_url': str,
6279 'concurrent_view_count': int,
6280 'channel': str,
6281 'uploader': str,
6282 'uploader_url': str,
6283 'uploader_id': str,
6284 'channel_is_verified': bool, # this will keep changing
6285 }
6286 }],
6287 'params': {'extract_flat': True, 'playlist_items': '1'},
6288 'playlist_mincount': 1
6289 }, {
6290 # Channel renderer metadata. Contains number of videos on the channel
6291 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6292 'info_dict': {
6293 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6294 'title': 'cole-dlp-test-acc - Channels',
6295 'channel': 'cole-dlp-test-acc',
6296 'description': 'test description',
6297 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6298 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6299 'tags': [],
6300 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6301 'uploader_id': '@coletdjnz',
6302 'uploader': 'cole-dlp-test-acc',
6303 },
6304 'playlist': [{
6305 'info_dict': {
6306 '_type': 'url',
6307 'ie_key': 'YoutubeTab',
6308 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6309 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6310 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6311 'title': 'PewDiePie',
6312 'channel': 'PewDiePie',
6313 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6314 'thumbnails': list,
6315 'channel_follower_count': int,
6316 'playlist_count': int,
6317 'uploader': 'PewDiePie',
6318 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6319 'uploader_id': '@PewDiePie',
6320 'channel_is_verified': True,
6321 }
6322 }],
6323 'params': {'extract_flat': True},
6324 }, {
6325 'url': 'https://www.youtube.com/@3blue1brown/about',
6326 'info_dict': {
6327 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6328 'tags': ['Mathematics'],
6329 'title': '3Blue1Brown - About',
6330 'channel_follower_count': int,
6331 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6332 'channel': '3Blue1Brown',
6333 'view_count': int,
6334 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6335 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6336 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6337 'uploader_id': '@3blue1brown',
6338 'uploader': '3Blue1Brown',
6339 'channel_is_verified': True,
6340 },
6341 'playlist_count': 0,
6342 }, {
6343 # Podcasts tab, with rich entry playlistRenderers
6344 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6345 'info_dict': {
6346 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6347 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6348 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6349 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6350 'title': '99 Percent Invisible - Podcasts',
6351 'uploader': '99 Percent Invisible',
6352 'channel_follower_count': int,
6353 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6354 'tags': [],
6355 'channel': '99 Percent Invisible',
6356 'uploader_id': '@99percentinvisiblepodcast',
6357 },
6358 'playlist_count': 1,
6359 }, {
6360 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6361 'url': 'https://www.youtube.com/@AHimitsu/releases',
6362 'info_dict': {
6363 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6364 'channel': 'A Himitsu',
6365 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6366 'title': 'A Himitsu - Releases',
6367 'uploader_id': '@AHimitsu',
6368 'uploader': 'A Himitsu',
6369 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6370 'tags': 'count:16',
6371 'description': 'I make music',
6372 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6373 'channel_follower_count': int,
6374 'channel_is_verified': True,
6375 },
6376 'playlist_mincount': 10,
6377 }, {
6378 # Playlist with only shorts, shown as reel renderers
6379 # FIXME: future: YouTube currently doesn't give continuation for this,
6380 # may do in future.
6381 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6382 'info_dict': {
6383 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6384 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6385 'view_count': int,
6386 'uploader_id': '@BangyShorts',
6387 'description': '',
6388 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6389 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6390 'channel': 'Bangy Shorts',
6391 'uploader': 'Bangy Shorts',
6392 'tags': [],
6393 'availability': 'public',
6394 'modified_date': '20230626',
6395 'title': 'Uploads from Bangy Shorts',
6396 },
6397 'playlist_mincount': 100,
6398 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6399 }]
6400
6401 @classmethod
6402 def suitable(cls, url):
6403 return False if YoutubeIE.suitable(url) else super().suitable(url)
6404
6405 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6406
6407 def _get_url_mobj(self, url):
6408 mobj = self._URL_RE.match(url).groupdict()
6409 mobj.update((k, '') for k, v in mobj.items() if v is None)
6410 return mobj
6411
6412 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6413 tab_name = (tab.get('title') or '').lower()
6414 tab_url = urljoin(base_url, traverse_obj(
6415 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6416
6417 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6418 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6419 if tab_id:
6420 return {
6421 'TAB_ID_SPONSORSHIPS': 'membership',
6422 }.get(tab_id, tab_id), tab_name
6423
6424 # Fallback to tab name if we cannot get the tab id.
6425 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6426 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6427 if tab_name:
6428 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6429 return {
6430 'home': 'featured',
6431 'live': 'streams',
6432 }.get(tab_name, tab_name), tab_name
6433
6434 def _has_tab(self, tabs, tab_id):
6435 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6436
6437 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6438 def _real_extract(self, url, smuggled_data):
6439 item_id = self._match_id(url)
6440 url = urllib.parse.urlunparse(
6441 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6442 compat_opts = self.get_param('compat_opts', [])
6443
6444 mobj = self._get_url_mobj(url)
6445 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6446 if is_channel and smuggled_data.get('is_music_url'):
6447 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6448 return self.url_result(
6449 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6450 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6451 mdata = self._extract_tab_endpoint(
6452 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6453 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6454 get_all=False, expected_type=str)
6455 if not murl:
6456 raise ExtractorError('Failed to resolve album to playlist')
6457 return self.url_result(murl, YoutubeTabIE)
6458 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6459 return self.url_result(
6460 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6461
6462 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6463 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6464 url = f'{pre}/videos{post}'
6465 if smuggled_data.get('is_music_url'):
6466 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6467
6468 # Handle both video/playlist URLs
6469 qs = parse_qs(url)
6470 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6471 if not video_id and mobj['not_channel'].startswith('watch'):
6472 if not playlist_id:
6473 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6474 raise ExtractorError('A video URL was given without video ID', expected=True)
6475 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6476 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6477 return self.url_result(
6478 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6479
6480 if not self._yes_playlist(playlist_id, video_id):
6481 return self.url_result(
6482 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6483
6484 data, ytcfg = self._extract_data(url, display_id)
6485
6486 # YouTube may provide a non-standard redirect to the regional channel
6487 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6488 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6489 redirect_url = traverse_obj(
6490 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6491 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6492 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6493 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6494 return self.url_result(redirect_url, YoutubeTabIE)
6495
6496 tabs, extra_tabs = self._extract_tab_renderers(data), []
6497 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6498 selected_tab = self._extract_selected_tab(tabs)
6499 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6500 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6501
6502 if not original_tab_id and selected_tab_name:
6503 self.to_screen('Downloading all uploads of the channel. '
6504 'To download only the videos in a specific tab, pass the tab\'s URL')
6505 if self._has_tab(tabs, 'streams'):
6506 extra_tabs.append(''.join((pre, '/streams', post)))
6507 if self._has_tab(tabs, 'shorts'):
6508 extra_tabs.append(''.join((pre, '/shorts', post)))
6509 # XXX: Members-only tab should also be extracted
6510
6511 if not extra_tabs and selected_tab_id != 'videos':
6512 # Channel does not have streams, shorts or videos tabs
6513 if item_id[:2] != 'UC':
6514 raise ExtractorError('This channel has no uploads', expected=True)
6515
6516 # Topic channels don't have /videos. Use the equivalent playlist instead
6517 pl_id = f'UU{item_id[2:]}'
6518 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6519 try:
6520 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6521 except ExtractorError:
6522 raise ExtractorError('This channel has no uploads', expected=True)
6523 else:
6524 item_id, url = pl_id, pl_url
6525 self.to_screen(
6526 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6527
6528 elif extra_tabs and selected_tab_id != 'videos':
6529 # When there are shorts/live tabs but not videos tab
6530 url, data = f'{pre}{post}', None
6531
6532 elif (original_tab_id or 'videos') != selected_tab_id:
6533 if original_tab_id == 'live':
6534 # Live tab should have redirected to the video
6535 # Except in the case the channel has an actual live tab
6536 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6537 raise UserNotLive(video_id=item_id)
6538 elif selected_tab_name:
6539 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6540
6541 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6542 url = f'{pre}{post}'
6543
6544 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6545 if 'no-youtube-unavailable-videos' not in compat_opts:
6546 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6547 self._extract_and_report_alerts(data, only_once=True)
6548
6549 tabs, entries = self._extract_tab_renderers(data), []
6550 if tabs:
6551 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6552 entries[0].update({
6553 'extractor_key': YoutubeTabIE.ie_key(),
6554 'extractor': YoutubeTabIE.IE_NAME,
6555 'webpage_url': url,
6556 })
6557 if self.get_param('playlist_items') == '0':
6558 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6559 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6560 entries.extend(map(self._real_extract, extra_tabs))
6561
6562 if len(entries) == 1:
6563 return entries[0]
6564 elif entries:
6565 metadata = self._extract_metadata_from_tabs(item_id, data)
6566 uploads_url = 'the Uploads (UU) playlist URL'
6567 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6568 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6569 self.to_screen(
6570 'Downloading as multiple playlists, separated by tabs. '
6571 f'To download as a single playlist instead, pass {uploads_url}')
6572 return self.playlist_result(entries, item_id, **metadata)
6573
6574 # Inline playlist
6575 playlist = traverse_obj(
6576 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6577 if playlist:
6578 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6579
6580 video_id = traverse_obj(
6581 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6582 if video_id:
6583 if tab != '/live': # live tab is expected to redirect to video
6584 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6585 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6586
6587 raise ExtractorError('Unable to recognize tab page')
6588
6589
6590 class YoutubePlaylistIE(InfoExtractor):
6591 IE_DESC = 'YouTube playlists'
6592 _VALID_URL = r'''(?x)(?:
6593 (?:https?://)?
6594 (?:\w+\.)?
6595 (?:
6596 (?:
6597 youtube(?:kids)?\.com|
6598 %(invidious)s
6599 )
6600 /.*?\?.*?\blist=
6601 )?
6602 (?P<id>%(playlist_id)s)
6603 )''' % {
6604 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6605 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6606 }
6607 IE_NAME = 'youtube:playlist'
6608 _TESTS = [{
6609 'note': 'issue #673',
6610 'url': 'PLBB231211A4F62143',
6611 'info_dict': {
6612 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6613 'id': 'PLBB231211A4F62143',
6614 'uploader': 'Wickman',
6615 'uploader_id': '@WickmanVT',
6616 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6617 'view_count': int,
6618 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6619 'modified_date': r're:\d{8}',
6620 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6621 'channel': 'Wickman',
6622 'tags': [],
6623 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6624 'availability': 'public',
6625 },
6626 'playlist_mincount': 29,
6627 }, {
6628 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6629 'info_dict': {
6630 'title': 'YDL_safe_search',
6631 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6632 },
6633 'playlist_count': 2,
6634 'skip': 'This playlist is private',
6635 }, {
6636 'note': 'embedded',
6637 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6638 'playlist_count': 4,
6639 'info_dict': {
6640 'title': 'JODA15',
6641 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6642 'uploader': 'milan',
6643 'uploader_id': '@milan5503',
6644 'description': '',
6645 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6646 'tags': [],
6647 'modified_date': '20140919',
6648 'view_count': int,
6649 'channel': 'milan',
6650 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6651 'uploader_url': 'https://www.youtube.com/@milan5503',
6652 'availability': 'public',
6653 },
6654 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6655 }, {
6656 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6657 'playlist_mincount': 455,
6658 'info_dict': {
6659 'title': '2018 Chinese New Singles (11/6 updated)',
6660 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6661 'uploader': 'LBK',
6662 'uploader_id': '@music_king',
6663 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6664 'channel': 'LBK',
6665 'view_count': int,
6666 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6667 'tags': [],
6668 'uploader_url': 'https://www.youtube.com/@music_king',
6669 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6670 'modified_date': r're:\d{8}',
6671 'availability': 'public',
6672 },
6673 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6674 }, {
6675 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6676 'only_matching': True,
6677 }, {
6678 # music album playlist
6679 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6680 'only_matching': True,
6681 }]
6682
6683 @classmethod
6684 def suitable(cls, url):
6685 if YoutubeTabIE.suitable(url):
6686 return False
6687 from ..utils import parse_qs
6688 qs = parse_qs(url)
6689 if qs.get('v', [None])[0]:
6690 return False
6691 return super().suitable(url)
6692
6693 def _real_extract(self, url):
6694 playlist_id = self._match_id(url)
6695 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6696 url = update_url_query(
6697 'https://www.youtube.com/playlist',
6698 parse_qs(url) or {'list': playlist_id})
6699 if is_music_url:
6700 url = smuggle_url(url, {'is_music_url': True})
6701 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6702
6703
6704 class YoutubeYtBeIE(InfoExtractor):
6705 IE_DESC = 'youtu.be'
6706 _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}
6707 _TESTS = [{
6708 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6709 'info_dict': {
6710 'id': 'yeWKywCrFtk',
6711 'ext': 'mp4',
6712 'title': 'Small Scale Baler and Braiding Rugs',
6713 'uploader': 'Backus-Page House Museum',
6714 'uploader_id': '@backuspagemuseum',
6715 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6716 'upload_date': '20161008',
6717 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6718 'categories': ['Nonprofits & Activism'],
6719 'tags': list,
6720 'like_count': int,
6721 'age_limit': 0,
6722 'playable_in_embed': True,
6723 'thumbnail': r're:^https?://.*\.webp',
6724 'channel': 'Backus-Page House Museum',
6725 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6726 'live_status': 'not_live',
6727 'view_count': int,
6728 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6729 'availability': 'public',
6730 'duration': 59,
6731 'comment_count': int,
6732 'channel_follower_count': int
6733 },
6734 'params': {
6735 'noplaylist': True,
6736 'skip_download': True,
6737 },
6738 }, {
6739 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6740 'only_matching': True,
6741 }]
6742
6743 def _real_extract(self, url):
6744 mobj = self._match_valid_url(url)
6745 video_id = mobj.group('id')
6746 playlist_id = mobj.group('playlist_id')
6747 return self.url_result(
6748 update_url_query('https://www.youtube.com/watch', {
6749 'v': video_id,
6750 'list': playlist_id,
6751 'feature': 'youtu.be',
6752 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6753
6754
6755 class YoutubeLivestreamEmbedIE(InfoExtractor):
6756 IE_DESC = 'YouTube livestream embeds'
6757 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6758 _TESTS = [{
6759 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6760 'only_matching': True,
6761 }]
6762
6763 def _real_extract(self, url):
6764 channel_id = self._match_id(url)
6765 return self.url_result(
6766 f'https://www.youtube.com/channel/{channel_id}/live',
6767 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6768
6769
6770 class YoutubeYtUserIE(InfoExtractor):
6771 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6772 IE_NAME = 'youtube:user'
6773 _VALID_URL = r'ytuser:(?P<id>.+)'
6774 _TESTS = [{
6775 'url': 'ytuser:phihag',
6776 'only_matching': True,
6777 }]
6778
6779 def _real_extract(self, url):
6780 user_id = self._match_id(url)
6781 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6782
6783
6784 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6785 IE_NAME = 'youtube:favorites'
6786 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6787 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6788 _LOGIN_REQUIRED = True
6789 _TESTS = [{
6790 'url': ':ytfav',
6791 'only_matching': True,
6792 }, {
6793 'url': ':ytfavorites',
6794 'only_matching': True,
6795 }]
6796
6797 def _real_extract(self, url):
6798 return self.url_result(
6799 'https://www.youtube.com/playlist?list=LL',
6800 ie=YoutubeTabIE.ie_key())
6801
6802
6803 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6804 IE_NAME = 'youtube:notif'
6805 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6806 _VALID_URL = r':ytnotif(?:ication)?s?'
6807 _LOGIN_REQUIRED = True
6808 _TESTS = [{
6809 'url': ':ytnotif',
6810 'only_matching': True,
6811 }, {
6812 'url': ':ytnotifications',
6813 'only_matching': True,
6814 }]
6815
6816 def _extract_notification_menu(self, response, continuation_list):
6817 notification_list = traverse_obj(
6818 response,
6819 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6820 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6821 expected_type=list) or []
6822 continuation_list[0] = None
6823 for item in notification_list:
6824 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6825 if entry:
6826 yield entry
6827 continuation = item.get('continuationItemRenderer')
6828 if continuation:
6829 continuation_list[0] = continuation
6830
6831 def _extract_notification_renderer(self, notification):
6832 video_id = traverse_obj(
6833 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6834 url = f'https://www.youtube.com/watch?v={video_id}'
6835 channel_id = None
6836 if not video_id:
6837 browse_ep = traverse_obj(
6838 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6839 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6840 post_id = self._search_regex(
6841 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6842 'post id', default=None)
6843 if not channel_id or not post_id:
6844 return
6845 # The direct /post url redirects to this in the browser
6846 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6847
6848 channel = traverse_obj(
6849 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6850 expected_type=str)
6851 notification_title = self._get_text(notification, 'shortMessage')
6852 if notification_title:
6853 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6854 # TODO: handle recommended videos
6855 title = self._search_regex(
6856 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6857 'video title', default=None)
6858 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6859 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6860 else None)
6861 return {
6862 '_type': 'url',
6863 'url': url,
6864 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6865 'video_id': video_id,
6866 'title': title,
6867 'channel_id': channel_id,
6868 'channel': channel,
6869 'uploader': channel,
6870 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6871 'timestamp': timestamp,
6872 }
6873
6874 def _notification_menu_entries(self, ytcfg):
6875 continuation_list = [None]
6876 response = None
6877 for page in itertools.count(1):
6878 ctoken = traverse_obj(
6879 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6880 response = self._extract_response(
6881 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6882 ep='notification/get_notification_menu', check_get_keys='actions',
6883 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6884 yield from self._extract_notification_menu(response, continuation_list)
6885 if not continuation_list[0]:
6886 break
6887
6888 def _real_extract(self, url):
6889 display_id = 'notifications'
6890 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6891 self._report_playlist_authcheck(ytcfg)
6892 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6893
6894
6895 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6896 IE_DESC = 'YouTube search'
6897 IE_NAME = 'youtube:search'
6898 _SEARCH_KEY = 'ytsearch'
6899 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6900 _TESTS = [{
6901 'url': 'ytsearch5:youtube-dl test video',
6902 'playlist_count': 5,
6903 'info_dict': {
6904 'id': 'youtube-dl test video',
6905 'title': 'youtube-dl test video',
6906 }
6907 }]
6908
6909
6910 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6911 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6912 _SEARCH_KEY = 'ytsearchdate'
6913 IE_DESC = 'YouTube search, newest videos first'
6914 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6915 _TESTS = [{
6916 'url': 'ytsearchdate5:youtube-dl test video',
6917 'playlist_count': 5,
6918 'info_dict': {
6919 'id': 'youtube-dl test video',
6920 'title': 'youtube-dl test video',
6921 }
6922 }]
6923
6924
6925 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6926 IE_DESC = 'YouTube search URLs with sorting and filter support'
6927 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6928 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6929 _TESTS = [{
6930 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6931 'playlist_mincount': 5,
6932 'info_dict': {
6933 'id': 'youtube-dl test video',
6934 'title': 'youtube-dl test video',
6935 }
6936 }, {
6937 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6938 'playlist_mincount': 5,
6939 'info_dict': {
6940 'id': 'python',
6941 'title': 'python',
6942 }
6943 }, {
6944 'url': 'https://www.youtube.com/results?search_query=%23cats',
6945 'playlist_mincount': 1,
6946 'info_dict': {
6947 'id': '#cats',
6948 'title': '#cats',
6949 # The test suite does not have support for nested playlists
6950 # 'entries': [{
6951 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6952 # 'title': '#cats',
6953 # }],
6954 },
6955 }, {
6956 # Channel results
6957 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6958 'info_dict': {
6959 'id': 'kurzgesagt',
6960 'title': 'kurzgesagt',
6961 },
6962 'playlist': [{
6963 'info_dict': {
6964 '_type': 'url',
6965 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6966 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6967 'ie_key': 'YoutubeTab',
6968 'channel': 'Kurzgesagt – In a Nutshell',
6969 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6970 'title': 'Kurzgesagt – In a Nutshell',
6971 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6972 # No longer available for search as it is set to the handle.
6973 # 'playlist_count': int,
6974 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6975 'thumbnails': list,
6976 'uploader_id': '@kurzgesagt',
6977 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
6978 'uploader': 'Kurzgesagt – In a Nutshell',
6979 'channel_is_verified': True,
6980 'channel_follower_count': int,
6981 }
6982 }],
6983 'params': {'extract_flat': True, 'playlist_items': '1'},
6984 'playlist_mincount': 1,
6985 }, {
6986 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
6987 'only_matching': True,
6988 }]
6989
6990 def _real_extract(self, url):
6991 qs = parse_qs(url)
6992 query = (qs.get('search_query') or qs.get('q'))[0]
6993 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
6994
6995
6996 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
6997 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
6998 IE_NAME = 'youtube:music:search_url'
6999 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7000 _TESTS = [{
7001 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7002 'playlist_count': 16,
7003 'info_dict': {
7004 'id': 'royalty free music',
7005 'title': 'royalty free music',
7006 }
7007 }, {
7008 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7009 'playlist_mincount': 30,
7010 'info_dict': {
7011 'id': 'royalty free music - songs',
7012 'title': 'royalty free music - songs',
7013 },
7014 'params': {'extract_flat': 'in_playlist'}
7015 }, {
7016 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7017 'playlist_mincount': 30,
7018 'info_dict': {
7019 'id': 'royalty free music - community playlists',
7020 'title': 'royalty free music - community playlists',
7021 },
7022 'params': {'extract_flat': 'in_playlist'}
7023 }]
7024
7025 _SECTIONS = {
7026 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7027 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7028 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7029 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7030 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7031 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7032 }
7033
7034 def _real_extract(self, url):
7035 qs = parse_qs(url)
7036 query = (qs.get('search_query') or qs.get('q'))[0]
7037 params = qs.get('sp', (None,))[0]
7038 if params:
7039 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7040 else:
7041 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
7042 params = self._SECTIONS.get(section)
7043 if not params:
7044 section = None
7045 title = join_nonempty(query, section, delim=' - ')
7046 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7047
7048
7049 class YoutubeFeedsInfoExtractor(InfoExtractor):
7050 """
7051 Base class for feed extractors
7052 Subclasses must re-define the _FEED_NAME property.
7053 """
7054 _LOGIN_REQUIRED = True
7055 _FEED_NAME = 'feeds'
7056
7057 def _real_initialize(self):
7058 YoutubeBaseInfoExtractor._check_login_required(self)
7059
7060 @classproperty
7061 def IE_NAME(self):
7062 return f'youtube:{self._FEED_NAME}'
7063
7064 def _real_extract(self, url):
7065 return self.url_result(
7066 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7067
7068
7069 class YoutubeWatchLaterIE(InfoExtractor):
7070 IE_NAME = 'youtube:watchlater'
7071 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7072 _VALID_URL = r':ytwatchlater'
7073 _TESTS = [{
7074 'url': ':ytwatchlater',
7075 'only_matching': True,
7076 }]
7077
7078 def _real_extract(self, url):
7079 return self.url_result(
7080 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7081
7082
7083 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7084 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7085 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7086 _FEED_NAME = 'recommended'
7087 _LOGIN_REQUIRED = False
7088 _TESTS = [{
7089 'url': ':ytrec',
7090 'only_matching': True,
7091 }, {
7092 'url': ':ytrecommended',
7093 'only_matching': True,
7094 }, {
7095 'url': 'https://youtube.com',
7096 'only_matching': True,
7097 }]
7098
7099
7100 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7101 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7102 _VALID_URL = r':ytsub(?:scription)?s?'
7103 _FEED_NAME = 'subscriptions'
7104 _TESTS = [{
7105 'url': ':ytsubs',
7106 'only_matching': True,
7107 }, {
7108 'url': ':ytsubscriptions',
7109 'only_matching': True,
7110 }]
7111
7112
7113 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7114 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7115 _VALID_URL = r':ythis(?:tory)?'
7116 _FEED_NAME = 'history'
7117 _TESTS = [{
7118 'url': ':ythistory',
7119 'only_matching': True,
7120 }]
7121
7122
7123 class YoutubeShortsAudioPivotIE(InfoExtractor):
7124 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7125 IE_NAME = 'youtube:shorts:pivot:audio'
7126 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7127 _TESTS = [{
7128 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7129 'only_matching': True,
7130 }]
7131
7132 @staticmethod
7133 def _generate_audio_pivot_params(video_id):
7134 """
7135 Generates sfv_audio_pivot browse params for this video id
7136 """
7137 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7138 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7139
7140 def _real_extract(self, url):
7141 video_id = self._match_id(url)
7142 return self.url_result(
7143 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7144 ie=YoutubeTabIE)
7145
7146
7147 class YoutubeTruncatedURLIE(InfoExtractor):
7148 IE_NAME = 'youtube:truncated_url'
7149 IE_DESC = False # Do not list
7150 _VALID_URL = r'''(?x)
7151 (?:https?://)?
7152 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7153 (?:watch\?(?:
7154 feature=[a-z_]+|
7155 annotation_id=annotation_[^&]+|
7156 x-yt-cl=[0-9]+|
7157 hl=[^&]*|
7158 t=[0-9]+
7159 )?
7160 |
7161 attribution_link\?a=[^&]+
7162 )
7163 $
7164 '''
7165
7166 _TESTS = [{
7167 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7168 'only_matching': True,
7169 }, {
7170 'url': 'https://www.youtube.com/watch?',
7171 'only_matching': True,
7172 }, {
7173 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7174 'only_matching': True,
7175 }, {
7176 'url': 'https://www.youtube.com/watch?feature=foo',
7177 'only_matching': True,
7178 }, {
7179 'url': 'https://www.youtube.com/watch?hl=en-GB',
7180 'only_matching': True,
7181 }, {
7182 'url': 'https://www.youtube.com/watch?t=2372',
7183 'only_matching': True,
7184 }]
7185
7186 def _real_extract(self, url):
7187 raise ExtractorError(
7188 'Did you forget to quote the URL? Remember that & is a meta '
7189 'character in most shells, so you want to put the URL in quotes, '
7190 'like youtube-dl '
7191 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7192 ' or simply youtube-dl BaW_jenozKc .',
7193 expected=True)
7194
7195
7196 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7197 IE_NAME = 'youtube:clip'
7198 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7199 _TESTS = [{
7200 # FIXME: Other metadata should be extracted from the clip, not from the base video
7201 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7202 'info_dict': {
7203 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7204 'ext': 'mp4',
7205 'section_start': 29.0,
7206 'section_end': 39.7,
7207 'duration': 10.7,
7208 'age_limit': 0,
7209 'availability': 'public',
7210 'categories': ['Gaming'],
7211 'channel': 'Scott The Woz',
7212 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7213 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7214 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7215 'like_count': int,
7216 'playable_in_embed': True,
7217 'tags': 'count:17',
7218 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7219 'title': 'Mobile Games on Console - Scott The Woz',
7220 'upload_date': '20210920',
7221 'uploader': 'Scott The Woz',
7222 'uploader_id': '@ScottTheWoz',
7223 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7224 'view_count': int,
7225 'live_status': 'not_live',
7226 'channel_follower_count': int,
7227 'chapters': 'count:20',
7228 'comment_count': int,
7229 'heatmap': 'count:100',
7230 }
7231 }]
7232
7233 def _real_extract(self, url):
7234 clip_id = self._match_id(url)
7235 _, data = self._extract_webpage(url, clip_id)
7236
7237 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7238 if not video_id:
7239 raise ExtractorError('Unable to find video ID')
7240
7241 clip_data = traverse_obj(data, (
7242 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7243 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7244 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7245 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7246
7247 return {
7248 '_type': 'url_transparent',
7249 'url': f'https://www.youtube.com/watch?v={video_id}',
7250 'ie_key': YoutubeIE.ie_key(),
7251 'id': clip_id,
7252 'section_start': int(clip_data['startTimeMs']) / 1000,
7253 'section_end': int(clip_data['endTimeMs']) / 1000,
7254 }
7255
7256
7257 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7258 IE_NAME = 'youtube:consent'
7259 IE_DESC = False # Do not list
7260 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7261 _TESTS = [{
7262 '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',
7263 'info_dict': {
7264 'id': 'qVv6vCqciTM',
7265 'ext': 'mp4',
7266 'age_limit': 0,
7267 'uploader_id': '@sana_natori',
7268 'comment_count': int,
7269 'chapters': 'count:13',
7270 'upload_date': '20221223',
7271 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7272 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7273 'uploader_url': 'https://www.youtube.com/@sana_natori',
7274 'like_count': int,
7275 'release_date': '20221223',
7276 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7277 '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】',
7278 'view_count': int,
7279 'playable_in_embed': True,
7280 'duration': 4438,
7281 'availability': 'public',
7282 'channel_follower_count': int,
7283 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7284 'categories': ['Entertainment'],
7285 'live_status': 'was_live',
7286 'release_timestamp': 1671793345,
7287 'channel': 'さなちゃんねる',
7288 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7289 'uploader': 'さなちゃんねる',
7290 'channel_is_verified': True,
7291 'heatmap': 'count:100',
7292 },
7293 'add_ie': ['Youtube'],
7294 'params': {'skip_download': 'Youtube'},
7295 }]
7296
7297 def _real_extract(self, url):
7298 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7299 if not redirect_url:
7300 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7301 return self.url_result(redirect_url)
7302
7303
7304 class YoutubeTruncatedIDIE(InfoExtractor):
7305 IE_NAME = 'youtube:truncated_id'
7306 IE_DESC = False # Do not list
7307 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7308
7309 _TESTS = [{
7310 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7311 'only_matching': True,
7312 }]
7313
7314 def _real_extract(self, url):
7315 video_id = self._match_id(url)
7316 raise ExtractorError(
7317 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7318 expected=True)