]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[extractor/youtube] Extract `heatmap` data (#7100)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime
6 import enum
7 import hashlib
8 import itertools
9 import json
10 import math
11 import os.path
12 import random
13 import re
14 import sys
15 import threading
16 import time
17 import traceback
18 import urllib.error
19 import urllib.parse
20
21 from .common import InfoExtractor, SearchInfoExtractor
22 from .openload import PhantomJSwrapper
23 from ..compat import functools
24 from ..jsinterp import JSInterpreter
25 from ..utils import (
26 NO_DEFAULT,
27 ExtractorError,
28 LazyList,
29 UserNotLive,
30 bug_reports_message,
31 classproperty,
32 clean_html,
33 datetime_from_str,
34 dict_get,
35 filter_dict,
36 float_or_none,
37 format_field,
38 get_first,
39 int_or_none,
40 is_html,
41 join_nonempty,
42 js_to_json,
43 mimetype2ext,
44 network_exceptions,
45 orderedSet,
46 parse_codecs,
47 parse_count,
48 parse_duration,
49 parse_iso8601,
50 parse_qs,
51 qualities,
52 remove_start,
53 smuggle_url,
54 str_or_none,
55 str_to_int,
56 strftime_or_none,
57 traverse_obj,
58 try_get,
59 unescapeHTML,
60 unified_strdate,
61 unified_timestamp,
62 unsmuggle_url,
63 update_url_query,
64 url_or_none,
65 urljoin,
66 variadic,
67 )
68
69 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
70 # any clients starting with _ cannot be explicitly requested by the user
71 INNERTUBE_CLIENTS = {
72 'web': {
73 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
74 'INNERTUBE_CONTEXT': {
75 'client': {
76 'clientName': 'WEB',
77 'clientVersion': '2.20220801.00.00',
78 }
79 },
80 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
81 },
82 'web_embedded': {
83 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
84 'INNERTUBE_CONTEXT': {
85 'client': {
86 'clientName': 'WEB_EMBEDDED_PLAYER',
87 'clientVersion': '1.20220731.00.00',
88 },
89 },
90 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
91 },
92 'web_music': {
93 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
94 'INNERTUBE_HOST': 'music.youtube.com',
95 'INNERTUBE_CONTEXT': {
96 'client': {
97 'clientName': 'WEB_REMIX',
98 'clientVersion': '1.20220727.01.00',
99 }
100 },
101 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
102 },
103 'web_creator': {
104 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
105 'INNERTUBE_CONTEXT': {
106 'client': {
107 'clientName': 'WEB_CREATOR',
108 'clientVersion': '1.20220726.00.00',
109 }
110 },
111 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
112 },
113 'android': {
114 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
115 'INNERTUBE_CONTEXT': {
116 'client': {
117 'clientName': 'ANDROID',
118 'clientVersion': '17.31.35',
119 'androidSdkVersion': 30,
120 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
121 }
122 },
123 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
124 'REQUIRE_JS_PLAYER': False
125 },
126 'android_embedded': {
127 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
128 'INNERTUBE_CONTEXT': {
129 'client': {
130 'clientName': 'ANDROID_EMBEDDED_PLAYER',
131 'clientVersion': '17.31.35',
132 'androidSdkVersion': 30,
133 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
134 },
135 },
136 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
137 'REQUIRE_JS_PLAYER': False
138 },
139 'android_music': {
140 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
141 'INNERTUBE_CONTEXT': {
142 'client': {
143 'clientName': 'ANDROID_MUSIC',
144 'clientVersion': '5.16.51',
145 'androidSdkVersion': 30,
146 'userAgent': 'com.google.android.apps.youtube.music/5.16.51 (Linux; U; Android 11) gzip'
147 }
148 },
149 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
150 'REQUIRE_JS_PLAYER': False
151 },
152 'android_creator': {
153 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
154 'INNERTUBE_CONTEXT': {
155 'client': {
156 'clientName': 'ANDROID_CREATOR',
157 'clientVersion': '22.30.100',
158 'androidSdkVersion': 30,
159 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
160 },
161 },
162 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
163 'REQUIRE_JS_PLAYER': False
164 },
165 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
166 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
167 'ios': {
168 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
169 'INNERTUBE_CONTEXT': {
170 'client': {
171 'clientName': 'IOS',
172 'clientVersion': '17.33.2',
173 'deviceModel': 'iPhone14,3',
174 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
175 }
176 },
177 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
178 'REQUIRE_JS_PLAYER': False
179 },
180 'ios_embedded': {
181 'INNERTUBE_CONTEXT': {
182 'client': {
183 'clientName': 'IOS_MESSAGES_EXTENSION',
184 'clientVersion': '17.33.2',
185 'deviceModel': 'iPhone14,3',
186 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
187 },
188 },
189 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
190 'REQUIRE_JS_PLAYER': False
191 },
192 'ios_music': {
193 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
194 'INNERTUBE_CONTEXT': {
195 'client': {
196 'clientName': 'IOS_MUSIC',
197 'clientVersion': '5.21',
198 'deviceModel': 'iPhone14,3',
199 'userAgent': 'com.google.ios.youtubemusic/5.21 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
200 },
201 },
202 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
203 'REQUIRE_JS_PLAYER': False
204 },
205 'ios_creator': {
206 'INNERTUBE_CONTEXT': {
207 'client': {
208 'clientName': 'IOS_CREATOR',
209 'clientVersion': '22.33.101',
210 'deviceModel': 'iPhone14,3',
211 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
212 },
213 },
214 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
215 'REQUIRE_JS_PLAYER': False
216 },
217 # mweb has 'ultralow' formats
218 # See: https://github.com/yt-dlp/yt-dlp/pull/557
219 'mweb': {
220 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
221 'INNERTUBE_CONTEXT': {
222 'client': {
223 'clientName': 'MWEB',
224 'clientVersion': '2.20220801.00.00',
225 }
226 },
227 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
228 },
229 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
230 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
231 'tv_embedded': {
232 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
233 'INNERTUBE_CONTEXT': {
234 'client': {
235 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
236 'clientVersion': '2.0',
237 },
238 },
239 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
240 },
241 }
242
243
244 def _split_innertube_client(client_name):
245 variant, *base = client_name.rsplit('.', 1)
246 if base:
247 return variant, base[0], variant
248 base, *variant = client_name.split('_', 1)
249 return client_name, base, variant[0] if variant else None
250
251
252 def short_client_name(client_name):
253 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
254 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
255
256
257 def build_innertube_clients():
258 THIRD_PARTY = {
259 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
260 }
261 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
262 priority = qualities(BASE_CLIENTS[::-1])
263
264 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
265 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
266 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
267 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
268 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
269
270 _, base_client, variant = _split_innertube_client(client)
271 ytcfg['priority'] = 10 * priority(base_client)
272
273 if not variant:
274 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
275 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
276 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
277 embedscreen['priority'] -= 3
278 elif variant == 'embedded':
279 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
280 ytcfg['priority'] -= 2
281 else:
282 ytcfg['priority'] -= 3
283
284
285 build_innertube_clients()
286
287
288 class BadgeType(enum.Enum):
289 AVAILABILITY_UNLISTED = enum.auto()
290 AVAILABILITY_PRIVATE = enum.auto()
291 AVAILABILITY_PUBLIC = enum.auto()
292 AVAILABILITY_PREMIUM = enum.auto()
293 AVAILABILITY_SUBSCRIPTION = enum.auto()
294 LIVE_NOW = enum.auto()
295
296
297 class YoutubeBaseInfoExtractor(InfoExtractor):
298 """Provide base functions for Youtube extractors"""
299
300 _RESERVED_NAMES = (
301 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
302 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
303 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
304 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
305
306 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
307
308 # _NETRC_MACHINE = 'youtube'
309
310 # If True it will raise an error if no login info is provided
311 _LOGIN_REQUIRED = False
312
313 _INVIDIOUS_SITES = (
314 # invidious-redirect websites
315 r'(?:www\.)?redirect\.invidious\.io',
316 r'(?:(?:www|dev)\.)?invidio\.us',
317 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
318 r'(?:www\.)?invidious\.pussthecat\.org',
319 r'(?:www\.)?invidious\.zee\.li',
320 r'(?:www\.)?invidious\.ethibox\.fr',
321 r'(?:www\.)?iv\.ggtyler\.dev',
322 r'(?:www\.)?inv\.vern\.i2p',
323 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
324 r'(?:www\.)?inv\.riverside\.rocks',
325 r'(?:www\.)?invidious\.silur\.me',
326 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
327 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
328 r'(?:www\.)?invidious\.slipfox\.xyz',
329 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
330 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
331 r'(?:www\.)?invidious\.tiekoetter\.com',
332 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
333 r'(?:www\.)?invidious\.nerdvpn\.de',
334 r'(?:www\.)?invidious\.weblibre\.org',
335 r'(?:www\.)?inv\.odyssey346\.dev',
336 r'(?:www\.)?invidious\.dhusch\.de',
337 r'(?:www\.)?iv\.melmac\.space',
338 r'(?:www\.)?watch\.thekitty\.zone',
339 r'(?:www\.)?invidious\.privacydev\.net',
340 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
341 r'(?:www\.)?invidious\.drivet\.xyz',
342 r'(?:www\.)?vid\.priv\.au',
343 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
344 r'(?:www\.)?inv\.vern\.cc',
345 r'(?:www\.)?invidious\.esmailelbob\.xyz',
346 r'(?:www\.)?invidious\.sethforprivacy\.com',
347 r'(?:www\.)?yt\.oelrichsgarcia\.de',
348 r'(?:www\.)?yt\.artemislena\.eu',
349 r'(?:www\.)?invidious\.flokinet\.to',
350 r'(?:www\.)?invidious\.baczek\.me',
351 r'(?:www\.)?y\.com\.sb',
352 r'(?:www\.)?invidious\.epicsite\.xyz',
353 r'(?:www\.)?invidious\.lidarshield\.cloud',
354 r'(?:www\.)?yt\.funami\.tech',
355 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
356 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
357 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
358 # youtube-dl invidious instances list
359 r'(?:(?:www|no)\.)?invidiou\.sh',
360 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
361 r'(?:www\.)?invidious\.kabi\.tk',
362 r'(?:www\.)?invidious\.mastodon\.host',
363 r'(?:www\.)?invidious\.zapashcanon\.fr',
364 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
365 r'(?:www\.)?invidious\.tinfoil-hat\.net',
366 r'(?:www\.)?invidious\.himiko\.cloud',
367 r'(?:www\.)?invidious\.reallyancient\.tech',
368 r'(?:www\.)?invidious\.tube',
369 r'(?:www\.)?invidiou\.site',
370 r'(?:www\.)?invidious\.site',
371 r'(?:www\.)?invidious\.xyz',
372 r'(?:www\.)?invidious\.nixnet\.xyz',
373 r'(?:www\.)?invidious\.048596\.xyz',
374 r'(?:www\.)?invidious\.drycat\.fr',
375 r'(?:www\.)?inv\.skyn3t\.in',
376 r'(?:www\.)?tube\.poal\.co',
377 r'(?:www\.)?tube\.connect\.cafe',
378 r'(?:www\.)?vid\.wxzm\.sx',
379 r'(?:www\.)?vid\.mint\.lgbt',
380 r'(?:www\.)?vid\.puffyan\.us',
381 r'(?:www\.)?yewtu\.be',
382 r'(?:www\.)?yt\.elukerio\.org',
383 r'(?:www\.)?yt\.lelux\.fi',
384 r'(?:www\.)?invidious\.ggc-project\.de',
385 r'(?:www\.)?yt\.maisputain\.ovh',
386 r'(?:www\.)?ytprivate\.com',
387 r'(?:www\.)?invidious\.13ad\.de',
388 r'(?:www\.)?invidious\.toot\.koeln',
389 r'(?:www\.)?invidious\.fdn\.fr',
390 r'(?:www\.)?watch\.nettohikari\.com',
391 r'(?:www\.)?invidious\.namazso\.eu',
392 r'(?:www\.)?invidious\.silkky\.cloud',
393 r'(?:www\.)?invidious\.exonip\.de',
394 r'(?:www\.)?invidious\.riverside\.rocks',
395 r'(?:www\.)?invidious\.blamefran\.net',
396 r'(?:www\.)?invidious\.moomoo\.de',
397 r'(?:www\.)?ytb\.trom\.tf',
398 r'(?:www\.)?yt\.cyberhost\.uk',
399 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
400 r'(?:www\.)?qklhadlycap4cnod\.onion',
401 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
402 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
403 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
404 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
405 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
406 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
407 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
408 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
409 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
410 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
411 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
412 r'(?:www\.)?piped\.kavin\.rocks',
413 r'(?:www\.)?piped\.tokhmi\.xyz',
414 r'(?:www\.)?piped\.syncpundit\.io',
415 r'(?:www\.)?piped\.mha\.fi',
416 r'(?:www\.)?watch\.whatever\.social',
417 r'(?:www\.)?piped\.garudalinux\.org',
418 r'(?:www\.)?piped\.rivo\.lol',
419 r'(?:www\.)?piped-libre\.kavin\.rocks',
420 r'(?:www\.)?yt\.jae\.fi',
421 r'(?:www\.)?piped\.mint\.lgbt',
422 r'(?:www\.)?il\.ax',
423 r'(?:www\.)?piped\.esmailelbob\.xyz',
424 r'(?:www\.)?piped\.projectsegfau\.lt',
425 r'(?:www\.)?piped\.privacydev\.net',
426 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
427 r'(?:www\.)?piped\.smnz\.de',
428 r'(?:www\.)?piped\.adminforge\.de',
429 r'(?:www\.)?watch\.whatevertinfoil\.de',
430 r'(?:www\.)?piped\.qdi\.fi',
431 r'(?:www\.)?piped\.video',
432 r'(?:www\.)?piped\.aeong\.one',
433 r'(?:www\.)?piped\.moomoo\.me',
434 r'(?:www\.)?piped\.chauvet\.pro',
435 r'(?:www\.)?watch\.leptons\.xyz',
436 r'(?:www\.)?pd\.vern\.cc',
437 r'(?:www\.)?piped\.hostux\.net',
438 r'(?:www\.)?piped\.lunar\.icu',
439 # Hyperpipe instances from https://hyperpipe.codeberg.page/
440 r'(?:www\.)?hyperpipe\.surge\.sh',
441 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
442 r'(?:www\.)?listen\.whatever\.social',
443 r'(?:www\.)?music\.adminforge\.de',
444 )
445
446 # extracted from account/account_menu ep
447 # XXX: These are the supported YouTube UI and API languages,
448 # which is slightly different from languages supported for translation in YouTube studio
449 _SUPPORTED_LANG_CODES = [
450 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
451 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
452 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
453 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
454 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
455 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
456 ]
457
458 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
459
460 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
461 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
462
463 def ucid_or_none(self, ucid):
464 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
465
466 def handle_or_none(self, handle):
467 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
468
469 def handle_from_url(self, url):
470 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
471 url, 'channel handle', default=None)
472
473 def ucid_from_url(self, url):
474 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
475 url, 'channel id', default=None)
476
477 @functools.cached_property
478 def _preferred_lang(self):
479 """
480 Returns a language code supported by YouTube for the user preferred language.
481 Returns None if no preferred language set.
482 """
483 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
484 if not preferred_lang:
485 return
486 if preferred_lang not in self._SUPPORTED_LANG_CODES:
487 raise ExtractorError(
488 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
489 expected=True)
490 elif preferred_lang != 'en':
491 self.report_warning(
492 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
493 return preferred_lang
494
495 def _initialize_consent(self):
496 cookies = self._get_cookies('https://www.youtube.com/')
497 if cookies.get('__Secure-3PSID'):
498 return
499 consent_id = None
500 consent = cookies.get('CONSENT')
501 if consent:
502 if 'YES' in consent.value:
503 return
504 consent_id = self._search_regex(
505 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
506 if not consent_id:
507 consent_id = random.randint(100, 999)
508 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
509
510 def _initialize_pref(self):
511 cookies = self._get_cookies('https://www.youtube.com/')
512 pref_cookie = cookies.get('PREF')
513 pref = {}
514 if pref_cookie:
515 try:
516 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
517 except ValueError:
518 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
519 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
520 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
521
522 def _real_initialize(self):
523 self._initialize_pref()
524 self._initialize_consent()
525 self._check_login_required()
526
527 def _check_login_required(self):
528 if self._LOGIN_REQUIRED and not self._cookies_passed:
529 self.raise_login_required('Login details are needed to download this content', method='cookies')
530
531 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
532 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
533
534 def _get_default_ytcfg(self, client='web'):
535 return copy.deepcopy(INNERTUBE_CLIENTS[client])
536
537 def _get_innertube_host(self, client='web'):
538 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
539
540 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
541 # try_get but with fallback to default ytcfg client values when present
542 _func = lambda y: try_get(y, getter, expected_type)
543 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
544
545 def _extract_client_name(self, ytcfg, default_client='web'):
546 return self._ytcfg_get_safe(
547 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
548 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
549
550 def _extract_client_version(self, ytcfg, default_client='web'):
551 return self._ytcfg_get_safe(
552 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
553 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
554
555 def _select_api_hostname(self, req_api_hostname, default_client=None):
556 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
557 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
558
559 def _extract_api_key(self, ytcfg=None, default_client='web'):
560 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
561
562 def _extract_context(self, ytcfg=None, default_client='web'):
563 context = get_first(
564 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
565 # Enforce language and tz for extraction
566 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
567 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
568 return context
569
570 _SAPISID = None
571
572 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
573 time_now = round(time.time())
574 if self._SAPISID is None:
575 yt_cookies = self._get_cookies('https://www.youtube.com')
576 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
577 # See: https://github.com/yt-dlp/yt-dlp/issues/393
578 sapisid_cookie = dict_get(
579 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
580 if sapisid_cookie and sapisid_cookie.value:
581 self._SAPISID = sapisid_cookie.value
582 self.write_debug('Extracted SAPISID cookie')
583 # SAPISID cookie is required if not already present
584 if not yt_cookies.get('SAPISID'):
585 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
586 self._set_cookie(
587 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
588 else:
589 self._SAPISID = False
590 if not self._SAPISID:
591 return None
592 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
593 sapisidhash = hashlib.sha1(
594 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
595 return f'SAPISIDHASH {time_now}_{sapisidhash}'
596
597 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
598 note='Downloading API JSON', errnote='Unable to download API page',
599 context=None, api_key=None, api_hostname=None, default_client='web'):
600
601 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
602 data.update(query)
603 real_headers = self.generate_api_headers(default_client=default_client)
604 real_headers.update({'content-type': 'application/json'})
605 if headers:
606 real_headers.update(headers)
607 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
608 or api_key or self._extract_api_key(default_client=default_client))
609 return self._download_json(
610 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
611 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
612 data=json.dumps(data).encode('utf8'), headers=real_headers,
613 query={'key': api_key, 'prettyPrint': 'false'})
614
615 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
616 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
617
618 @staticmethod
619 def _extract_session_index(*data):
620 """
621 Index of current account in account list.
622 See: https://github.com/yt-dlp/yt-dlp/pull/519
623 """
624 for ytcfg in data:
625 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
626 if session_index is not None:
627 return session_index
628
629 # Deprecated?
630 def _extract_identity_token(self, ytcfg=None, webpage=None):
631 if ytcfg:
632 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
633 if token:
634 return token
635 if webpage:
636 return self._search_regex(
637 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
638 'identity token', default=None, fatal=False)
639
640 @staticmethod
641 def _extract_account_syncid(*args):
642 """
643 Extract syncId required to download private playlists of secondary channels
644 @params response and/or ytcfg
645 """
646 for data in args:
647 # ytcfg includes channel_syncid if on secondary channel
648 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
649 if delegated_sid:
650 return delegated_sid
651 sync_ids = (try_get(
652 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
653 lambda x: x['DATASYNC_ID']), str) or '').split('||')
654 if len(sync_ids) >= 2 and sync_ids[1]:
655 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
656 # and just "user_syncid||" for primary channel. We only want the channel_syncid
657 return sync_ids[0]
658
659 @staticmethod
660 def _extract_visitor_data(*args):
661 """
662 Extracts visitorData from an API response or ytcfg
663 Appears to be used to track session state
664 """
665 return get_first(
666 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
667 expected_type=str)
668
669 @functools.cached_property
670 def is_authenticated(self):
671 return bool(self._generate_sapisidhash_header())
672
673 def extract_ytcfg(self, video_id, webpage):
674 if not webpage:
675 return {}
676 return self._parse_json(
677 self._search_regex(
678 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
679 default='{}'), video_id, fatal=False) or {}
680
681 def generate_api_headers(
682 self, *, ytcfg=None, account_syncid=None, session_index=None,
683 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
684
685 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
686 headers = {
687 'X-YouTube-Client-Name': str(
688 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
689 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
690 'Origin': origin,
691 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
692 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
693 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
694 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
695 }
696 if session_index is None:
697 session_index = self._extract_session_index(ytcfg)
698 if account_syncid or session_index is not None:
699 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
700
701 auth = self._generate_sapisidhash_header(origin)
702 if auth is not None:
703 headers['Authorization'] = auth
704 headers['X-Origin'] = origin
705 return filter_dict(headers)
706
707 def _download_ytcfg(self, client, video_id):
708 url = {
709 'web': 'https://www.youtube.com',
710 'web_music': 'https://music.youtube.com',
711 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
712 }.get(client)
713 if not url:
714 return {}
715 webpage = self._download_webpage(
716 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
717 return self.extract_ytcfg(video_id, webpage) or {}
718
719 @staticmethod
720 def _build_api_continuation_query(continuation, ctp=None):
721 query = {
722 'continuation': continuation
723 }
724 # TODO: Inconsistency with clickTrackingParams.
725 # Currently we have a fixed ctp contained within context (from ytcfg)
726 # and a ctp in root query for continuation.
727 if ctp:
728 query['clickTracking'] = {'clickTrackingParams': ctp}
729 return query
730
731 @classmethod
732 def _extract_next_continuation_data(cls, renderer):
733 next_continuation = try_get(
734 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
735 lambda x: x['continuation']['reloadContinuationData']), dict)
736 if not next_continuation:
737 return
738 continuation = next_continuation.get('continuation')
739 if not continuation:
740 return
741 ctp = next_continuation.get('clickTrackingParams')
742 return cls._build_api_continuation_query(continuation, ctp)
743
744 @classmethod
745 def _extract_continuation_ep_data(cls, continuation_ep: dict):
746 if isinstance(continuation_ep, dict):
747 continuation = try_get(
748 continuation_ep, lambda x: x['continuationCommand']['token'], str)
749 if not continuation:
750 return
751 ctp = continuation_ep.get('clickTrackingParams')
752 return cls._build_api_continuation_query(continuation, ctp)
753
754 @classmethod
755 def _extract_continuation(cls, renderer):
756 next_continuation = cls._extract_next_continuation_data(renderer)
757 if next_continuation:
758 return next_continuation
759
760 return traverse_obj(renderer, (
761 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
762 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
763 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
764
765 @classmethod
766 def _extract_alerts(cls, data):
767 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
768 if not isinstance(alert_dict, dict):
769 continue
770 for alert in alert_dict.values():
771 alert_type = alert.get('type')
772 if not alert_type:
773 continue
774 message = cls._get_text(alert, 'text')
775 if message:
776 yield alert_type, message
777
778 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
779 errors, warnings = [], []
780 for alert_type, alert_message in alerts:
781 if alert_type.lower() == 'error' and fatal:
782 errors.append([alert_type, alert_message])
783 elif alert_message not in self._IGNORED_WARNINGS:
784 warnings.append([alert_type, alert_message])
785
786 for alert_type, alert_message in (warnings + errors[:-1]):
787 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
788 if errors:
789 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
790
791 def _extract_and_report_alerts(self, data, *args, **kwargs):
792 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
793
794 def _extract_badges(self, renderer: dict):
795 privacy_icon_map = {
796 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
797 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
798 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC
799 }
800
801 badge_style_map = {
802 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
803 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
804 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW
805 }
806
807 label_map = {
808 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
809 'private': BadgeType.AVAILABILITY_PRIVATE,
810 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
811 'live': BadgeType.LIVE_NOW,
812 'premium': BadgeType.AVAILABILITY_PREMIUM
813 }
814
815 badges = []
816 for badge in traverse_obj(renderer, ('badges', ..., 'metadataBadgeRenderer')):
817 badge_type = (
818 privacy_icon_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
819 or badge_style_map.get(traverse_obj(badge, 'style'))
820 )
821 if badge_type:
822 badges.append({'type': badge_type})
823 continue
824
825 # fallback, won't work in some languages
826 label = traverse_obj(badge, 'label', expected_type=str, default='')
827 for match, label_badge_type in label_map.items():
828 if match in label.lower():
829 badges.append({'type': badge_type})
830 continue
831
832 return badges
833
834 @staticmethod
835 def _has_badge(badges, badge_type):
836 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
837
838 @staticmethod
839 def _get_text(data, *path_list, max_runs=None):
840 for path in path_list or [None]:
841 if path is None:
842 obj = [data]
843 else:
844 obj = traverse_obj(data, path, default=[])
845 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
846 obj = [obj]
847 for item in obj:
848 text = try_get(item, lambda x: x['simpleText'], str)
849 if text:
850 return text
851 runs = try_get(item, lambda x: x['runs'], list) or []
852 if not runs and isinstance(item, list):
853 runs = item
854
855 runs = runs[:min(len(runs), max_runs or len(runs))]
856 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
857 if text:
858 return text
859
860 def _get_count(self, data, *path_list):
861 count_text = self._get_text(data, *path_list) or ''
862 count = parse_count(count_text)
863 if count is None:
864 count = str_to_int(
865 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
866 return count
867
868 @staticmethod
869 def _extract_thumbnails(data, *path_list):
870 """
871 Extract thumbnails from thumbnails dict
872 @param path_list: path list to level that contains 'thumbnails' key
873 """
874 thumbnails = []
875 for path in path_list or [()]:
876 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
877 thumbnail_url = url_or_none(thumbnail.get('url'))
878 if not thumbnail_url:
879 continue
880 # Sometimes youtube gives a wrong thumbnail URL. See:
881 # https://github.com/yt-dlp/yt-dlp/issues/233
882 # https://github.com/ytdl-org/youtube-dl/issues/28023
883 if 'maxresdefault' in thumbnail_url:
884 thumbnail_url = thumbnail_url.split('?')[0]
885 thumbnails.append({
886 'url': thumbnail_url,
887 'height': int_or_none(thumbnail.get('height')),
888 'width': int_or_none(thumbnail.get('width')),
889 })
890 return thumbnails
891
892 @staticmethod
893 def extract_relative_time(relative_time_text):
894 """
895 Extracts a relative time from string and converts to dt object
896 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
897 """
898 mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
899 if mobj:
900 start = mobj.group('start')
901 if start:
902 return datetime_from_str(start)
903 try:
904 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
905 except ValueError:
906 return None
907
908 def _parse_time_text(self, text):
909 if not text:
910 return
911 dt = self.extract_relative_time(text)
912 timestamp = None
913 if isinstance(dt, datetime.datetime):
914 timestamp = calendar.timegm(dt.timetuple())
915
916 if timestamp is None:
917 timestamp = (
918 unified_timestamp(text) or unified_timestamp(
919 self._search_regex(
920 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
921 text.lower(), 'time text', default=None)))
922
923 if text and timestamp is None and self._preferred_lang in (None, 'en'):
924 self.report_warning(
925 f'Cannot parse localized time text "{text}"', only_once=True)
926 return timestamp
927
928 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
929 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
930 default_client='web'):
931 for retry in self.RetryManager():
932 try:
933 response = self._call_api(
934 ep=ep, fatal=True, headers=headers,
935 video_id=item_id, query=query, note=note,
936 context=self._extract_context(ytcfg, default_client),
937 api_key=self._extract_api_key(ytcfg, default_client),
938 api_hostname=api_hostname, default_client=default_client)
939 except ExtractorError as e:
940 if not isinstance(e.cause, network_exceptions):
941 return self._error_or_warning(e, fatal=fatal)
942 elif not isinstance(e.cause, urllib.error.HTTPError):
943 retry.error = e
944 continue
945
946 first_bytes = e.cause.read(512)
947 if not is_html(first_bytes):
948 yt_error = try_get(
949 self._parse_json(
950 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
951 lambda x: x['error']['message'], str)
952 if yt_error:
953 self._report_alerts([('ERROR', yt_error)], fatal=False)
954 # Downloading page may result in intermittent 5xx HTTP error
955 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
956 # We also want to catch all other network exceptions since errors in later pages can be troublesome
957 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
958 if e.cause.code not in (403, 429):
959 retry.error = e
960 continue
961 return self._error_or_warning(e, fatal=fatal)
962
963 try:
964 self._extract_and_report_alerts(response, only_once=True)
965 except ExtractorError as e:
966 # YouTube servers may return errors we want to retry on in a 200 OK response
967 # See: https://github.com/yt-dlp/yt-dlp/issues/839
968 if 'unknown error' in e.msg.lower():
969 retry.error = e
970 continue
971 return self._error_or_warning(e, fatal=fatal)
972 # Youtube sometimes sends incomplete data
973 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
974 if not traverse_obj(response, *variadic(check_get_keys)):
975 retry.error = ExtractorError('Incomplete data received', expected=True)
976 continue
977
978 return response
979
980 @staticmethod
981 def is_music_url(url):
982 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
983
984 def _extract_video(self, renderer):
985 video_id = renderer.get('videoId')
986
987 reel_header_renderer = traverse_obj(renderer, (
988 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
989 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
990
991 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
992 description = self._get_text(renderer, 'descriptionSnippet')
993
994 duration = int_or_none(renderer.get('lengthSeconds'))
995 if duration is None:
996 duration = parse_duration(self._get_text(
997 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
998 if duration is None:
999 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1000 duration = parse_duration(self._search_regex(
1001 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1002 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1003 video_id, default=None, group='duration'))
1004
1005 channel_id = traverse_obj(
1006 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1007 expected_type=str, get_all=False)
1008 if not channel_id:
1009 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1010
1011 channel_id = self.ucid_or_none(channel_id)
1012
1013 overlay_style = traverse_obj(
1014 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1015 get_all=False, expected_type=str)
1016 badges = self._extract_badges(renderer)
1017
1018 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1019 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1020 expected_type=str)) or ''
1021 url = f'https://www.youtube.com/watch?v={video_id}'
1022 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1023 url = f'https://www.youtube.com/shorts/{video_id}'
1024
1025 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1026 or self._get_text(reel_header_renderer, 'timestampText') or '')
1027 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1028
1029 live_status = (
1030 'is_upcoming' if scheduled_timestamp is not None
1031 else 'was_live' if 'streamed' in time_text.lower()
1032 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1033 else None)
1034
1035 # videoInfo is a string like '50K views • 10 years ago'.
1036 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1037 view_count = (0 if 'no views' in view_count_text.lower()
1038 else self._get_count({'simpleText': view_count_text}))
1039 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1040
1041 return {
1042 '_type': 'url',
1043 'ie_key': YoutubeIE.ie_key(),
1044 'id': video_id,
1045 'url': url,
1046 'title': title,
1047 'description': description,
1048 'duration': duration,
1049 'channel_id': channel_id,
1050 'channel': (self._get_text(renderer, 'ownerText', 'shortBylineText')
1051 or self._get_text(reel_header_renderer, 'channelTitleText')),
1052 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1053 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1054 'timestamp': (self._parse_time_text(time_text)
1055 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1056 else None),
1057 'release_timestamp': scheduled_timestamp,
1058 'availability':
1059 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1060 else self._availability(
1061 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1062 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1063 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1064 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1065 view_count_field: view_count,
1066 'live_status': live_status
1067 }
1068
1069
1070 class YoutubeIE(YoutubeBaseInfoExtractor):
1071 IE_DESC = 'YouTube'
1072 _VALID_URL = r"""(?x)^
1073 (
1074 (?:https?://|//) # http(s):// or protocol-independent URL
1075 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1076 (?:www\.)?deturl\.com/www\.youtube\.com|
1077 (?:www\.)?pwnyoutube\.com|
1078 (?:www\.)?hooktube\.com|
1079 (?:www\.)?yourepeat\.com|
1080 tube\.majestyc\.net|
1081 %(invidious)s|
1082 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1083 (?:.*?\#/)? # handle anchor (#/) redirect urls
1084 (?: # the various things that can precede the ID:
1085 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1086 |(?: # or the v= param in all its forms
1087 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1088 (?:\?|\#!?) # the params delimiter ? or # or #!
1089 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1090 v=
1091 )
1092 ))
1093 |(?:
1094 youtu\.be| # just youtu.be/xxxx
1095 vid\.plus| # or vid.plus/xxxx
1096 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1097 %(invidious)s
1098 )/
1099 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1100 )
1101 )? # all until now is optional -> you can pass the naked ID
1102 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1103 (?(1).+)? # if we found the ID, everything can follow
1104 (?:\#|$)""" % {
1105 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1106 }
1107 _EMBED_REGEX = [
1108 r'''(?x)
1109 (?:
1110 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1111 data-video-url=|
1112 <embed[^>]+?src=|
1113 embedSWF\(?:\s*|
1114 <object[^>]+data=|
1115 new\s+SWFObject\(
1116 )
1117 (["\'])
1118 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1119 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1120 \1''',
1121 # https://wordpress.org/plugins/lazy-load-for-videos/
1122 r'''(?xs)
1123 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1124 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1125 ]
1126 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1127
1128 _PLAYER_INFO_RE = (
1129 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1130 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$',
1131 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1132 )
1133 _formats = {
1134 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1135 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1136 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1137 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1138 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1139 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1140 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1141 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1142 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1143 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1144 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1145 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1146 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1147 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1148 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1149 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1150 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1151 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1152
1153
1154 # 3D videos
1155 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1156 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1157 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1158 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1159 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1160 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1161 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1162
1163 # Apple HTTP Live Streaming
1164 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1165 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1166 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1167 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1168 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1169 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1170 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1171 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1172
1173 # DASH mp4 video
1174 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1175 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1176 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1177 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1178 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1179 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1180 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1181 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1182 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1183 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1184 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1185 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1186
1187 # Dash mp4 audio
1188 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1189 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1190 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1191 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1192 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1193 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1194 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1195
1196 # Dash webm
1197 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1198 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1199 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1200 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1201 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1202 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1203 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1204 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1205 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1206 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1207 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1208 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1209 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1210 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1211 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1212 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1213 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1214 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1215 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1216 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1217 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1218 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1219
1220 # Dash webm audio
1221 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1222 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1223
1224 # Dash webm audio with opus inside
1225 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1226 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1227 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1228
1229 # RTMP (unnamed)
1230 '_rtmp': {'protocol': 'rtmp'},
1231
1232 # av01 video only formats sometimes served with "unknown" codecs
1233 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1234 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1235 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1236 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1237 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1238 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1239 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1240 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1241 }
1242 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1243
1244 _GEO_BYPASS = False
1245
1246 IE_NAME = 'youtube'
1247 _TESTS = [
1248 {
1249 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1250 'info_dict': {
1251 'id': 'BaW_jenozKc',
1252 'ext': 'mp4',
1253 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1254 'channel': 'Philipp Hagemeister',
1255 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1256 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1257 'upload_date': '20121002',
1258 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1259 'categories': ['Science & Technology'],
1260 'tags': ['youtube-dl'],
1261 'duration': 10,
1262 'view_count': int,
1263 'like_count': int,
1264 'availability': 'public',
1265 'playable_in_embed': True,
1266 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1267 'live_status': 'not_live',
1268 'age_limit': 0,
1269 'start_time': 1,
1270 'end_time': 9,
1271 'comment_count': int,
1272 'channel_follower_count': int,
1273 'uploader': 'Philipp Hagemeister',
1274 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1275 'uploader_id': '@PhilippHagemeister',
1276 'heatmap': 'count:100',
1277 }
1278 },
1279 {
1280 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1281 'note': 'Embed-only video (#1746)',
1282 'info_dict': {
1283 'id': 'yZIXLfi8CZQ',
1284 'ext': 'mp4',
1285 'upload_date': '20120608',
1286 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1287 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1288 'age_limit': 18,
1289 },
1290 'skip': 'Private video',
1291 },
1292 {
1293 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1294 'note': 'Use the first video ID in the URL',
1295 'info_dict': {
1296 'id': 'BaW_jenozKc',
1297 'ext': 'mp4',
1298 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1299 'channel': 'Philipp Hagemeister',
1300 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1301 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1302 'upload_date': '20121002',
1303 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1304 'categories': ['Science & Technology'],
1305 'tags': ['youtube-dl'],
1306 'duration': 10,
1307 'view_count': int,
1308 'like_count': int,
1309 'availability': 'public',
1310 'playable_in_embed': True,
1311 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1312 'live_status': 'not_live',
1313 'age_limit': 0,
1314 'comment_count': int,
1315 'channel_follower_count': int,
1316 'uploader': 'Philipp Hagemeister',
1317 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1318 'uploader_id': '@PhilippHagemeister',
1319 },
1320 'params': {
1321 'skip_download': True,
1322 },
1323 },
1324 {
1325 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1326 'note': '256k DASH audio (format 141) via DASH manifest',
1327 'info_dict': {
1328 'id': 'a9LDPn-MO4I',
1329 'ext': 'm4a',
1330 'upload_date': '20121002',
1331 'description': '',
1332 'title': 'UHDTV TEST 8K VIDEO.mp4'
1333 },
1334 'params': {
1335 'youtube_include_dash_manifest': True,
1336 'format': '141',
1337 },
1338 'skip': 'format 141 not served anymore',
1339 },
1340 # DASH manifest with encrypted signature
1341 {
1342 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1343 'info_dict': {
1344 'id': 'IB3lcPjvWLA',
1345 'ext': 'm4a',
1346 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1347 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1348 'duration': 244,
1349 'upload_date': '20131011',
1350 'abr': 129.495,
1351 'like_count': int,
1352 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1353 'playable_in_embed': True,
1354 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1355 'view_count': int,
1356 'track': 'The Spark',
1357 'live_status': 'not_live',
1358 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1359 'channel': 'Afrojack',
1360 'tags': 'count:19',
1361 'availability': 'public',
1362 'categories': ['Music'],
1363 'age_limit': 0,
1364 'alt_title': 'The Spark',
1365 'channel_follower_count': int,
1366 'uploader': 'Afrojack',
1367 'uploader_url': 'https://www.youtube.com/@Afrojack',
1368 'uploader_id': '@Afrojack',
1369 },
1370 'params': {
1371 'youtube_include_dash_manifest': True,
1372 'format': '141/bestaudio[ext=m4a]',
1373 },
1374 },
1375 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1376 {
1377 'note': 'Embed allowed age-gate video',
1378 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1379 'info_dict': {
1380 'id': 'HtVdAasjOgU',
1381 'ext': 'mp4',
1382 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1383 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1384 'duration': 142,
1385 'upload_date': '20140605',
1386 'age_limit': 18,
1387 'categories': ['Gaming'],
1388 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1389 'availability': 'needs_auth',
1390 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1391 'like_count': int,
1392 'channel': 'The Witcher',
1393 'live_status': 'not_live',
1394 'tags': 'count:17',
1395 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1396 'playable_in_embed': True,
1397 'view_count': int,
1398 'channel_follower_count': int,
1399 'uploader': 'The Witcher',
1400 'uploader_url': 'https://www.youtube.com/@thewitcher',
1401 'uploader_id': '@thewitcher',
1402 },
1403 },
1404 {
1405 'note': 'Age-gate video with embed allowed in public site',
1406 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1407 'info_dict': {
1408 'id': 'HsUATh_Nc2U',
1409 'ext': 'mp4',
1410 'title': 'Godzilla 2 (Official Video)',
1411 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1412 'upload_date': '20200408',
1413 'age_limit': 18,
1414 'availability': 'needs_auth',
1415 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1416 'channel': 'FlyingKitty',
1417 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1418 'view_count': int,
1419 'categories': ['Entertainment'],
1420 'live_status': 'not_live',
1421 'tags': ['Flyingkitty', 'godzilla 2'],
1422 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1423 'like_count': int,
1424 'duration': 177,
1425 'playable_in_embed': True,
1426 'channel_follower_count': int,
1427 'uploader': 'FlyingKitty',
1428 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1429 'uploader_id': '@FlyingKitty900',
1430 'comment_count': int,
1431 },
1432 },
1433 {
1434 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1435 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1436 'info_dict': {
1437 'id': 'Tq92D6wQ1mg',
1438 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1439 'ext': 'mp4',
1440 'upload_date': '20191228',
1441 'description': 'md5:17eccca93a786d51bc67646756894066',
1442 'age_limit': 18,
1443 'like_count': int,
1444 'availability': 'needs_auth',
1445 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1446 'view_count': int,
1447 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1448 'channel': 'Projekt Melody',
1449 'live_status': 'not_live',
1450 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1451 'playable_in_embed': True,
1452 'categories': ['Entertainment'],
1453 'duration': 106,
1454 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1455 'comment_count': int,
1456 'channel_follower_count': int,
1457 'uploader': 'Projekt Melody',
1458 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1459 'uploader_id': '@ProjektMelody',
1460 },
1461 },
1462 {
1463 'note': 'Non-Agegated non-embeddable video',
1464 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1465 'info_dict': {
1466 'id': 'MeJVWBSsPAY',
1467 'ext': 'mp4',
1468 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1469 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1470 'upload_date': '20130730',
1471 'track': 'Such mich find mich',
1472 'age_limit': 0,
1473 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1474 'like_count': int,
1475 'playable_in_embed': False,
1476 'creator': 'OOMPH!',
1477 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1478 'view_count': int,
1479 'alt_title': 'Such mich find mich',
1480 'duration': 210,
1481 'channel': 'Herr Lurik',
1482 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1483 'categories': ['Music'],
1484 'availability': 'public',
1485 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1486 'live_status': 'not_live',
1487 'artist': 'OOMPH!',
1488 'channel_follower_count': int,
1489 'uploader': 'Herr Lurik',
1490 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1491 'uploader_id': '@HerrLurik',
1492 },
1493 },
1494 {
1495 'note': 'Non-bypassable age-gated video',
1496 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1497 'only_matching': True,
1498 },
1499 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1500 # YouTube Red ad is not captured for creator
1501 {
1502 'url': '__2ABJjxzNo',
1503 'info_dict': {
1504 'id': '__2ABJjxzNo',
1505 'ext': 'mp4',
1506 'duration': 266,
1507 'upload_date': '20100430',
1508 'creator': 'deadmau5',
1509 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1510 'title': 'Deadmau5 - Some Chords (HD)',
1511 'alt_title': 'Some Chords',
1512 'availability': 'public',
1513 'tags': 'count:14',
1514 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1515 'view_count': int,
1516 'live_status': 'not_live',
1517 'channel': 'deadmau5',
1518 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1519 'like_count': int,
1520 'track': 'Some Chords',
1521 'artist': 'deadmau5',
1522 'playable_in_embed': True,
1523 'age_limit': 0,
1524 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1525 'categories': ['Music'],
1526 'album': 'Some Chords',
1527 'channel_follower_count': int,
1528 'uploader': 'deadmau5',
1529 'uploader_url': 'https://www.youtube.com/@deadmau5',
1530 'uploader_id': '@deadmau5',
1531 },
1532 'expected_warnings': [
1533 'DASH manifest missing',
1534 ]
1535 },
1536 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1537 {
1538 'url': 'lqQg6PlCWgI',
1539 'info_dict': {
1540 'id': 'lqQg6PlCWgI',
1541 'ext': 'mp4',
1542 'duration': 6085,
1543 'upload_date': '20150827',
1544 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1545 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1546 'like_count': int,
1547 'release_timestamp': 1343767800,
1548 'playable_in_embed': True,
1549 'categories': ['Sports'],
1550 'release_date': '20120731',
1551 'channel': 'Olympics',
1552 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1553 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1554 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1555 'age_limit': 0,
1556 'availability': 'public',
1557 'live_status': 'was_live',
1558 'view_count': int,
1559 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1560 'channel_follower_count': int,
1561 'uploader': 'Olympics',
1562 'uploader_url': 'https://www.youtube.com/@Olympics',
1563 'uploader_id': '@Olympics',
1564 },
1565 'params': {
1566 'skip_download': 'requires avconv',
1567 }
1568 },
1569 # Non-square pixels
1570 {
1571 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1572 'info_dict': {
1573 'id': '_b-2C3KPAM0',
1574 'ext': 'mp4',
1575 'stretched_ratio': 16 / 9.,
1576 'duration': 85,
1577 'upload_date': '20110310',
1578 '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',
1579 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1580 'playable_in_embed': True,
1581 'channel': '孫ᄋᄅ',
1582 'age_limit': 0,
1583 'tags': 'count:11',
1584 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1585 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1586 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1587 'view_count': int,
1588 'categories': ['People & Blogs'],
1589 'like_count': int,
1590 'live_status': 'not_live',
1591 'availability': 'unlisted',
1592 'comment_count': int,
1593 'channel_follower_count': int,
1594 'uploader': '孫ᄋᄅ',
1595 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1596 'uploader_id': '@AllenMeow',
1597 },
1598 },
1599 # url_encoded_fmt_stream_map is empty string
1600 {
1601 'url': 'qEJwOuvDf7I',
1602 'info_dict': {
1603 'id': 'qEJwOuvDf7I',
1604 'ext': 'webm',
1605 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1606 'description': '',
1607 'upload_date': '20150404',
1608 },
1609 'params': {
1610 'skip_download': 'requires avconv',
1611 },
1612 'skip': 'This live event has ended.',
1613 },
1614 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1615 {
1616 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1617 'info_dict': {
1618 'id': 'FIl7x6_3R5Y',
1619 'ext': 'webm',
1620 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1621 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1622 'duration': 220,
1623 'upload_date': '20150625',
1624 'formats': 'mincount:31',
1625 },
1626 'skip': 'not actual anymore',
1627 },
1628 # DASH manifest with segment_list
1629 {
1630 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1631 'md5': '8ce563a1d667b599d21064e982ab9e31',
1632 'info_dict': {
1633 'id': 'CsmdDsKjzN8',
1634 'ext': 'mp4',
1635 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1636 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1637 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1638 },
1639 'params': {
1640 'youtube_include_dash_manifest': True,
1641 'format': '135', # bestvideo
1642 },
1643 'skip': 'This live event has ended.',
1644 },
1645 {
1646 # Multifeed videos (multiple cameras), URL can be of any Camera
1647 # TODO: fix multifeed titles
1648 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1649 'info_dict': {
1650 'id': 'zaPI8MvL8pg',
1651 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1652 'description': 'md5:563ccbc698b39298481ca3c571169519',
1653 },
1654 'playlist': [{
1655 'info_dict': {
1656 'id': 'j5yGuxZ8lLU',
1657 'ext': 'mp4',
1658 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1659 'description': 'md5:563ccbc698b39298481ca3c571169519',
1660 'duration': 10120,
1661 'channel_follower_count': int,
1662 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1663 'availability': 'public',
1664 'playable_in_embed': True,
1665 'upload_date': '20131105',
1666 'categories': ['Gaming'],
1667 'live_status': 'was_live',
1668 'tags': 'count:24',
1669 'release_timestamp': 1383701910,
1670 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1671 'comment_count': int,
1672 'age_limit': 0,
1673 'like_count': int,
1674 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1675 'channel': 'WiiLikeToPlay',
1676 'view_count': int,
1677 'release_date': '20131106',
1678 'uploader': 'WiiLikeToPlay',
1679 'uploader_id': '@WLTP',
1680 'uploader_url': 'https://www.youtube.com/@WLTP',
1681 },
1682 }, {
1683 'info_dict': {
1684 'id': 'zaPI8MvL8pg',
1685 'ext': 'mp4',
1686 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1687 'availability': 'public',
1688 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1689 'channel': 'WiiLikeToPlay',
1690 'channel_follower_count': int,
1691 'description': 'md5:563ccbc698b39298481ca3c571169519',
1692 'duration': 10108,
1693 'age_limit': 0,
1694 'like_count': int,
1695 'tags': 'count:24',
1696 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1697 'release_timestamp': 1383701915,
1698 'comment_count': int,
1699 'upload_date': '20131105',
1700 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1701 'release_date': '20131106',
1702 'playable_in_embed': True,
1703 'live_status': 'was_live',
1704 'categories': ['Gaming'],
1705 'view_count': int,
1706 'uploader': 'WiiLikeToPlay',
1707 'uploader_id': '@WLTP',
1708 'uploader_url': 'https://www.youtube.com/@WLTP',
1709 },
1710 }, {
1711 'info_dict': {
1712 'id': 'R7r3vfO7Hao',
1713 'ext': 'mp4',
1714 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1715 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1716 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1717 'like_count': int,
1718 'availability': 'public',
1719 'playable_in_embed': True,
1720 'upload_date': '20131105',
1721 'description': 'md5:563ccbc698b39298481ca3c571169519',
1722 'channel_follower_count': int,
1723 'tags': 'count:24',
1724 'release_date': '20131106',
1725 'comment_count': int,
1726 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1727 'channel': 'WiiLikeToPlay',
1728 'categories': ['Gaming'],
1729 'release_timestamp': 1383701914,
1730 'live_status': 'was_live',
1731 'age_limit': 0,
1732 'duration': 10128,
1733 'view_count': int,
1734 'uploader': 'WiiLikeToPlay',
1735 'uploader_id': '@WLTP',
1736 'uploader_url': 'https://www.youtube.com/@WLTP',
1737 },
1738 }],
1739 'params': {'skip_download': True},
1740 },
1741 {
1742 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1743 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1744 'info_dict': {
1745 'id': 'gVfLd0zydlo',
1746 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1747 },
1748 'playlist_count': 2,
1749 'skip': 'Not multifeed anymore',
1750 },
1751 {
1752 'url': 'https://vid.plus/FlRa-iH7PGw',
1753 'only_matching': True,
1754 },
1755 {
1756 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1757 'only_matching': True,
1758 },
1759 {
1760 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1761 # Also tests cut-off URL expansion in video description (see
1762 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1763 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1764 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1765 'info_dict': {
1766 'id': 'lsguqyKfVQg',
1767 'ext': 'mp4',
1768 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1769 'alt_title': 'Dark Walk',
1770 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1771 'duration': 133,
1772 'upload_date': '20151119',
1773 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1774 'track': 'Dark Walk',
1775 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1776 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1777 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1778 'categories': ['Film & Animation'],
1779 'view_count': int,
1780 'live_status': 'not_live',
1781 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1782 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1783 'tags': 'count:13',
1784 'availability': 'public',
1785 'channel': 'IronSoulElf',
1786 'playable_in_embed': True,
1787 'like_count': int,
1788 'age_limit': 0,
1789 'channel_follower_count': int
1790 },
1791 'params': {
1792 'skip_download': True,
1793 },
1794 },
1795 {
1796 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1797 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1798 'only_matching': True,
1799 },
1800 {
1801 # Video with yt:stretch=17:0
1802 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1803 'info_dict': {
1804 'id': 'Q39EVAstoRM',
1805 'ext': 'mp4',
1806 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1807 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1808 'upload_date': '20151107',
1809 },
1810 'params': {
1811 'skip_download': True,
1812 },
1813 'skip': 'This video does not exist.',
1814 },
1815 {
1816 # Video with incomplete 'yt:stretch=16:'
1817 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1818 'only_matching': True,
1819 },
1820 {
1821 # Video licensed under Creative Commons
1822 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1823 'info_dict': {
1824 'id': 'M4gD1WSo5mA',
1825 'ext': 'mp4',
1826 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1827 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1828 'duration': 721,
1829 'upload_date': '20150128',
1830 'license': 'Creative Commons Attribution license (reuse allowed)',
1831 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1832 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1833 'like_count': int,
1834 'age_limit': 0,
1835 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1836 'channel': 'The Berkman Klein Center for Internet & Society',
1837 'availability': 'public',
1838 'view_count': int,
1839 'categories': ['Education'],
1840 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1841 'live_status': 'not_live',
1842 'playable_in_embed': True,
1843 'channel_follower_count': int,
1844 'chapters': list,
1845 'uploader': 'The Berkman Klein Center for Internet & Society',
1846 'uploader_id': '@BKCHarvard',
1847 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1848 },
1849 'params': {
1850 'skip_download': True,
1851 },
1852 },
1853 {
1854 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1855 'info_dict': {
1856 'id': 'eQcmzGIKrzg',
1857 'ext': 'mp4',
1858 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1859 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1860 'duration': 4060,
1861 'upload_date': '20151120',
1862 'license': 'Creative Commons Attribution license (reuse allowed)',
1863 'playable_in_embed': True,
1864 'tags': 'count:12',
1865 'like_count': int,
1866 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1867 'age_limit': 0,
1868 'availability': 'public',
1869 'categories': ['News & Politics'],
1870 'channel': 'Bernie Sanders',
1871 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1872 'view_count': int,
1873 'live_status': 'not_live',
1874 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1875 'comment_count': int,
1876 'channel_follower_count': int,
1877 'chapters': list,
1878 'uploader': 'Bernie Sanders',
1879 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1880 'uploader_id': '@BernieSanders',
1881 },
1882 'params': {
1883 'skip_download': True,
1884 },
1885 },
1886 {
1887 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1888 'only_matching': True,
1889 },
1890 {
1891 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1892 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1893 'only_matching': True,
1894 },
1895 {
1896 # Rental video preview
1897 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1898 'info_dict': {
1899 'id': 'uGpuVWrhIzE',
1900 'ext': 'mp4',
1901 'title': 'Piku - Trailer',
1902 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1903 'upload_date': '20150811',
1904 'license': 'Standard YouTube License',
1905 },
1906 'params': {
1907 'skip_download': True,
1908 },
1909 'skip': 'This video is not available.',
1910 },
1911 {
1912 # YouTube Red video with episode data
1913 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1914 'info_dict': {
1915 'id': 'iqKdEhx-dD4',
1916 'ext': 'mp4',
1917 'title': 'Isolation - Mind Field (Ep 1)',
1918 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1919 'duration': 2085,
1920 'upload_date': '20170118',
1921 'series': 'Mind Field',
1922 'season_number': 1,
1923 'episode_number': 1,
1924 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1925 'tags': 'count:12',
1926 'view_count': int,
1927 'availability': 'public',
1928 'age_limit': 0,
1929 'channel': 'Vsauce',
1930 'episode': 'Episode 1',
1931 'categories': ['Entertainment'],
1932 'season': 'Season 1',
1933 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1934 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1935 'like_count': int,
1936 'playable_in_embed': True,
1937 'live_status': 'not_live',
1938 'channel_follower_count': int,
1939 'uploader': 'Vsauce',
1940 'uploader_url': 'https://www.youtube.com/@Vsauce',
1941 'uploader_id': '@Vsauce',
1942 },
1943 'params': {
1944 'skip_download': True,
1945 },
1946 'expected_warnings': [
1947 'Skipping DASH manifest',
1948 ],
1949 },
1950 {
1951 # The following content has been identified by the YouTube community
1952 # as inappropriate or offensive to some audiences.
1953 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1954 'info_dict': {
1955 'id': '6SJNVb0GnPI',
1956 'ext': 'mp4',
1957 'title': 'Race Differences in Intelligence',
1958 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1959 'duration': 965,
1960 'upload_date': '20140124',
1961 },
1962 'params': {
1963 'skip_download': True,
1964 },
1965 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1966 },
1967 {
1968 # itag 212
1969 'url': '1t24XAntNCY',
1970 'only_matching': True,
1971 },
1972 {
1973 # geo restricted to JP
1974 'url': 'sJL6WA-aGkQ',
1975 'only_matching': True,
1976 },
1977 {
1978 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1979 'only_matching': True,
1980 },
1981 {
1982 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1983 'only_matching': True,
1984 },
1985 {
1986 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1987 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1988 'only_matching': True,
1989 },
1990 {
1991 # DRM protected
1992 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1993 'only_matching': True,
1994 },
1995 {
1996 # Video with unsupported adaptive stream type formats
1997 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1998 'info_dict': {
1999 'id': 'Z4Vy8R84T1U',
2000 'ext': 'mp4',
2001 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2002 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2003 'duration': 433,
2004 'upload_date': '20130923',
2005 'formats': 'maxcount:10',
2006 },
2007 'params': {
2008 'skip_download': True,
2009 'youtube_include_dash_manifest': False,
2010 },
2011 'skip': 'not actual anymore',
2012 },
2013 {
2014 # Youtube Music Auto-generated description
2015 # TODO: fix metadata extraction
2016 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2017 'info_dict': {
2018 'id': 'MgNrAu2pzNs',
2019 'ext': 'mp4',
2020 'title': 'Voyeur Girl',
2021 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2022 'upload_date': '20190312',
2023 'artist': 'Stephen',
2024 'track': 'Voyeur Girl',
2025 'album': 'it\'s too much love to know my dear',
2026 'release_date': '20190313',
2027 'release_year': 2019,
2028 'alt_title': 'Voyeur Girl',
2029 'view_count': int,
2030 'playable_in_embed': True,
2031 'like_count': int,
2032 'categories': ['Music'],
2033 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2034 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2035 'uploader': 'Stephen',
2036 'availability': 'public',
2037 'creator': 'Stephen',
2038 'duration': 169,
2039 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2040 'age_limit': 0,
2041 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2042 'tags': 'count:11',
2043 'live_status': 'not_live',
2044 'channel_follower_count': int
2045 },
2046 'params': {
2047 'skip_download': True,
2048 },
2049 },
2050 {
2051 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2052 'only_matching': True,
2053 },
2054 {
2055 # invalid -> valid video id redirection
2056 'url': 'DJztXj2GPfl',
2057 'info_dict': {
2058 'id': 'DJztXj2GPfk',
2059 'ext': 'mp4',
2060 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2061 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2062 'upload_date': '20090125',
2063 'artist': 'Panjabi MC',
2064 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2065 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2066 },
2067 'params': {
2068 'skip_download': True,
2069 },
2070 'skip': 'Video unavailable',
2071 },
2072 {
2073 # empty description results in an empty string
2074 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2075 'info_dict': {
2076 'id': 'x41yOUIvK2k',
2077 'ext': 'mp4',
2078 'title': 'IMG 3456',
2079 'description': '',
2080 'upload_date': '20170613',
2081 'view_count': int,
2082 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2083 'like_count': int,
2084 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2085 'tags': [],
2086 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2087 'availability': 'public',
2088 'age_limit': 0,
2089 'categories': ['Pets & Animals'],
2090 'duration': 7,
2091 'playable_in_embed': True,
2092 'live_status': 'not_live',
2093 'channel': 'l\'Or Vert asbl',
2094 'channel_follower_count': int,
2095 'uploader': 'l\'Or Vert asbl',
2096 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2097 'uploader_id': '@ElevageOrVert',
2098 },
2099 'params': {
2100 'skip_download': True,
2101 },
2102 },
2103 {
2104 # with '};' inside yt initial data (see [1])
2105 # see [2] for an example with '};' inside ytInitialPlayerResponse
2106 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2107 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2108 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2109 'info_dict': {
2110 'id': 'CHqg6qOn4no',
2111 'ext': 'mp4',
2112 'title': 'Part 77 Sort a list of simple types in c#',
2113 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2114 'upload_date': '20130831',
2115 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2116 'like_count': int,
2117 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2118 'live_status': 'not_live',
2119 'categories': ['Education'],
2120 'availability': 'public',
2121 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2122 'tags': 'count:12',
2123 'playable_in_embed': True,
2124 'age_limit': 0,
2125 'view_count': int,
2126 'duration': 522,
2127 'channel': 'kudvenkat',
2128 'comment_count': int,
2129 'channel_follower_count': int,
2130 'chapters': list,
2131 'uploader': 'kudvenkat',
2132 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2133 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2134 },
2135 'params': {
2136 'skip_download': True,
2137 },
2138 },
2139 {
2140 # another example of '};' in ytInitialData
2141 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2142 'only_matching': True,
2143 },
2144 {
2145 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2146 'only_matching': True,
2147 },
2148 {
2149 # https://github.com/ytdl-org/youtube-dl/pull/28094
2150 'url': 'OtqTfy26tG0',
2151 'info_dict': {
2152 'id': 'OtqTfy26tG0',
2153 'ext': 'mp4',
2154 'title': 'Burn Out',
2155 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2156 'upload_date': '20141120',
2157 'artist': 'The Cinematic Orchestra',
2158 'track': 'Burn Out',
2159 'album': 'Every Day',
2160 'like_count': int,
2161 'live_status': 'not_live',
2162 'alt_title': 'Burn Out',
2163 'duration': 614,
2164 'age_limit': 0,
2165 'view_count': int,
2166 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2167 'creator': 'The Cinematic Orchestra',
2168 'channel': 'The Cinematic Orchestra',
2169 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2170 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2171 'availability': 'public',
2172 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2173 'categories': ['Music'],
2174 'playable_in_embed': True,
2175 'channel_follower_count': int,
2176 'uploader': 'The Cinematic Orchestra',
2177 'comment_count': int,
2178 },
2179 'params': {
2180 'skip_download': True,
2181 },
2182 },
2183 {
2184 # controversial video, only works with bpctr when authenticated with cookies
2185 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2186 'only_matching': True,
2187 },
2188 {
2189 # controversial video, requires bpctr/contentCheckOk
2190 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2191 'info_dict': {
2192 'id': 'SZJvDhaSDnc',
2193 'ext': 'mp4',
2194 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2195 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2196 'upload_date': '20140716',
2197 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2198 'duration': 170,
2199 'categories': ['News & Politics'],
2200 'view_count': int,
2201 'channel': 'CBS Mornings',
2202 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2203 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2204 'age_limit': 18,
2205 'availability': 'needs_auth',
2206 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2207 'like_count': int,
2208 'live_status': 'not_live',
2209 'playable_in_embed': True,
2210 'channel_follower_count': int,
2211 'uploader': 'CBS Mornings',
2212 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2213 'uploader_id': '@CBSMornings',
2214 }
2215 },
2216 {
2217 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2218 'url': 'cBvYw8_A0vQ',
2219 'info_dict': {
2220 'id': 'cBvYw8_A0vQ',
2221 'ext': 'mp4',
2222 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2223 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2224 'upload_date': '20201120',
2225 'duration': 1456,
2226 'categories': ['Travel & Events'],
2227 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2228 'view_count': int,
2229 'channel': 'Walk around Japan',
2230 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2231 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2232 'age_limit': 0,
2233 'availability': 'public',
2234 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2235 'live_status': 'not_live',
2236 'playable_in_embed': True,
2237 'channel_follower_count': int,
2238 'uploader': 'Walk around Japan',
2239 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2240 'uploader_id': '@walkaroundjapan7124',
2241 },
2242 'params': {
2243 'skip_download': True,
2244 },
2245 }, {
2246 # Has multiple audio streams
2247 'url': 'WaOKSUlf4TM',
2248 'only_matching': True
2249 }, {
2250 # Requires Premium: has format 141 when requested using YTM url
2251 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2252 'only_matching': True
2253 }, {
2254 # multiple subtitles with same lang_code
2255 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2256 'only_matching': True,
2257 }, {
2258 # Force use android client fallback
2259 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2260 'info_dict': {
2261 'id': 'YOelRv7fMxY',
2262 'title': 'DIGGING A SECRET TUNNEL Part 1',
2263 'ext': '3gp',
2264 'upload_date': '20210624',
2265 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2266 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2267 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2268 'duration': 596,
2269 'categories': ['Entertainment'],
2270 'view_count': int,
2271 'channel': 'colinfurze',
2272 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2273 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2274 'age_limit': 0,
2275 'availability': 'public',
2276 'like_count': int,
2277 'live_status': 'not_live',
2278 'playable_in_embed': True,
2279 'channel_follower_count': int,
2280 'chapters': list,
2281 'uploader': 'colinfurze',
2282 'uploader_url': 'https://www.youtube.com/@colinfurze',
2283 'uploader_id': '@colinfurze',
2284 },
2285 'params': {
2286 'format': '17', # 3gp format available on android
2287 'extractor_args': {'youtube': {'player_client': ['android']}},
2288 },
2289 },
2290 {
2291 # Skip download of additional client configs (remix client config in this case)
2292 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2293 'only_matching': True,
2294 'params': {
2295 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2296 },
2297 }, {
2298 # shorts
2299 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2300 'only_matching': True,
2301 }, {
2302 'note': 'Storyboards',
2303 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2304 'info_dict': {
2305 'id': '5KLPxDtMqe8',
2306 'ext': 'mhtml',
2307 'format_id': 'sb0',
2308 'title': 'Your Brain is Plastic',
2309 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2310 'upload_date': '20140324',
2311 'like_count': int,
2312 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2313 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2314 'view_count': int,
2315 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2316 'playable_in_embed': True,
2317 'tags': 'count:12',
2318 'availability': 'public',
2319 'channel': 'SciShow',
2320 'live_status': 'not_live',
2321 'duration': 248,
2322 'categories': ['Education'],
2323 'age_limit': 0,
2324 'channel_follower_count': int,
2325 'chapters': list,
2326 'uploader': 'SciShow',
2327 'uploader_url': 'https://www.youtube.com/@SciShow',
2328 'uploader_id': '@SciShow',
2329 }, 'params': {'format': 'mhtml', 'skip_download': True}
2330 }, {
2331 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2332 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2333 'info_dict': {
2334 'id': '2NUZ8W2llS4',
2335 'ext': 'mp4',
2336 'title': 'The NP that test your phone performance 🙂',
2337 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2338 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2339 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2340 'duration': 21,
2341 'view_count': int,
2342 'age_limit': 0,
2343 'categories': ['Gaming'],
2344 'tags': 'count:23',
2345 'playable_in_embed': True,
2346 'live_status': 'not_live',
2347 'upload_date': '20220103',
2348 'like_count': int,
2349 'availability': 'public',
2350 'channel': 'Leon Nguyen',
2351 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2352 'comment_count': int,
2353 'channel_follower_count': int,
2354 'uploader': 'Leon Nguyen',
2355 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2356 'uploader_id': '@LeonNguyen',
2357 }
2358 }, {
2359 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2360 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2361 'info_dict': {
2362 'id': '2NUZ8W2llS4',
2363 'ext': 'mp4',
2364 'title': 'The NP that test your phone performance 🙂',
2365 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2366 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2367 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2368 'duration': 21,
2369 'view_count': int,
2370 'age_limit': 0,
2371 'categories': ['Gaming'],
2372 'tags': 'count:23',
2373 'playable_in_embed': True,
2374 'live_status': 'not_live',
2375 'upload_date': '20220102',
2376 'like_count': int,
2377 'availability': 'public',
2378 'channel': 'Leon Nguyen',
2379 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2380 'comment_count': int,
2381 'channel_follower_count': int,
2382 'uploader': 'Leon Nguyen',
2383 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2384 'uploader_id': '@LeonNguyen',
2385 },
2386 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2387 }, {
2388 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2389 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2390 'info_dict': {
2391 'id': 'mzZzzBU6lrM',
2392 'ext': 'mp4',
2393 'title': 'I Met GeorgeNotFound In Real Life...',
2394 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2395 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2396 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2397 'duration': 955,
2398 'view_count': int,
2399 'age_limit': 0,
2400 'categories': ['Entertainment'],
2401 'tags': 'count:26',
2402 'playable_in_embed': True,
2403 'live_status': 'not_live',
2404 'release_timestamp': 1641172509,
2405 'release_date': '20220103',
2406 'upload_date': '20220103',
2407 'like_count': int,
2408 'availability': 'public',
2409 'channel': 'Quackity',
2410 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2411 'channel_follower_count': int,
2412 'uploader': 'Quackity',
2413 'uploader_id': '@Quackity',
2414 'uploader_url': 'https://www.youtube.com/@Quackity',
2415 }
2416 },
2417 { # continuous livestream. Microformat upload date should be preferred.
2418 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2419 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2420 'info_dict': {
2421 'id': 'kgx4WGK0oNU',
2422 '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}',
2423 'ext': 'mp4',
2424 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2425 'availability': 'public',
2426 'age_limit': 0,
2427 'release_timestamp': 1637975704,
2428 'upload_date': '20210619',
2429 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2430 'live_status': 'is_live',
2431 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2432 'channel': 'Abao in Tokyo',
2433 'channel_follower_count': int,
2434 'release_date': '20211127',
2435 'tags': 'count:39',
2436 'categories': ['People & Blogs'],
2437 'like_count': int,
2438 'view_count': int,
2439 'playable_in_embed': True,
2440 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2441 'concurrent_view_count': int,
2442 'uploader': 'Abao in Tokyo',
2443 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2444 'uploader_id': '@abaointokyo',
2445 },
2446 'params': {'skip_download': True}
2447 }, {
2448 # Story. Requires specific player params to work.
2449 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2450 'info_dict': {
2451 'id': 'vv8qTUWmulI',
2452 'ext': 'mp4',
2453 'availability': 'unlisted',
2454 'view_count': int,
2455 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2456 'upload_date': '20220526',
2457 'categories': ['Education'],
2458 'title': 'Story',
2459 'channel': 'IT\'S HISTORY',
2460 'description': '',
2461 'duration': 12,
2462 'playable_in_embed': True,
2463 'age_limit': 0,
2464 'live_status': 'not_live',
2465 'tags': [],
2466 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2467 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2468 },
2469 'skip': 'stories get removed after some period of time',
2470 }, {
2471 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2472 'info_dict': {
2473 'id': 'tjjjtzRLHvA',
2474 'ext': 'mp4',
2475 'title': 'ハッシュタグ無し };if window.ytcsi',
2476 'upload_date': '20220323',
2477 'like_count': int,
2478 'availability': 'unlisted',
2479 'channel': 'Lesmiscore',
2480 'thumbnail': r're:^https?://.*\.jpg',
2481 'age_limit': 0,
2482 'categories': ['Music'],
2483 'view_count': int,
2484 'description': '',
2485 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2486 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2487 'live_status': 'not_live',
2488 'playable_in_embed': True,
2489 'channel_follower_count': int,
2490 'duration': 6,
2491 'tags': [],
2492 'uploader_id': '@lesmiscore',
2493 'uploader': 'Lesmiscore',
2494 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2495 }
2496 }, {
2497 # Prefer primary title+description language metadata by default
2498 # Do not prefer translated description if primary is empty
2499 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2500 'info_dict': {
2501 'id': 'el3E4MbxRqQ',
2502 'ext': 'mp4',
2503 'title': 'dlp test video 2 - primary sv no desc',
2504 'description': '',
2505 'channel': 'cole-dlp-test-acc',
2506 'tags': [],
2507 'view_count': int,
2508 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2509 'like_count': int,
2510 'playable_in_embed': True,
2511 'availability': 'unlisted',
2512 'thumbnail': r're:^https?://.*\.jpg',
2513 'age_limit': 0,
2514 'duration': 5,
2515 'live_status': 'not_live',
2516 'upload_date': '20220908',
2517 'categories': ['People & Blogs'],
2518 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2519 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2520 'uploader_id': '@coletdjnz',
2521 'uploader': 'cole-dlp-test-acc',
2522 },
2523 'params': {'skip_download': True}
2524 }, {
2525 # Extractor argument: prefer translated title+description
2526 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2527 'info_dict': {
2528 'id': 'gHKT4uU8Zng',
2529 'ext': 'mp4',
2530 'channel': 'cole-dlp-test-acc',
2531 'tags': [],
2532 'duration': 5,
2533 'live_status': 'not_live',
2534 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2535 'upload_date': '20220728',
2536 'view_count': int,
2537 'categories': ['People & Blogs'],
2538 'thumbnail': r're:^https?://.*\.jpg',
2539 'title': 'dlp test video title translated (fr)',
2540 'availability': 'public',
2541 'age_limit': 0,
2542 'description': 'dlp test video description translated (fr)',
2543 'playable_in_embed': True,
2544 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2545 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2546 'uploader_id': '@coletdjnz',
2547 'uploader': 'cole-dlp-test-acc',
2548 },
2549 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2550 'expected_warnings': [r'Preferring "fr" translated fields'],
2551 }, {
2552 'note': '6 channel audio',
2553 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2554 'only_matching': True,
2555 }, {
2556 'note': 'Multiple HLS formats with same itag',
2557 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2558 'info_dict': {
2559 'id': 'kX3nB4PpJko',
2560 'ext': 'mp4',
2561 'categories': ['Entertainment'],
2562 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2563 'live_status': 'not_live',
2564 'duration': 937,
2565 'channel_follower_count': int,
2566 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2567 'title': 'Last To Take Hand Off Jet, Keeps It!',
2568 'channel': 'MrBeast',
2569 'playable_in_embed': True,
2570 'view_count': int,
2571 'upload_date': '20221112',
2572 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2573 'age_limit': 0,
2574 'availability': 'public',
2575 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2576 'like_count': int,
2577 'tags': [],
2578 'uploader': 'MrBeast',
2579 'uploader_url': 'https://www.youtube.com/@MrBeast',
2580 'uploader_id': '@MrBeast',
2581 },
2582 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2583 }, {
2584 'note': 'Audio formats with Dynamic Range Compression',
2585 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2586 'info_dict': {
2587 'id': 'Tq92D6wQ1mg',
2588 'ext': 'webm',
2589 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2590 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2591 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2592 'channel_follower_count': int,
2593 'description': 'md5:17eccca93a786d51bc67646756894066',
2594 'upload_date': '20191228',
2595 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2596 'playable_in_embed': True,
2597 'like_count': int,
2598 'categories': ['Entertainment'],
2599 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2600 'age_limit': 18,
2601 'channel': 'Projekt Melody',
2602 'view_count': int,
2603 'availability': 'needs_auth',
2604 'comment_count': int,
2605 'live_status': 'not_live',
2606 'duration': 106,
2607 'uploader': 'Projekt Melody',
2608 'uploader_id': '@ProjektMelody',
2609 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2610 },
2611 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2612 },
2613 {
2614 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2615 'info_dict': {
2616 'id': 'qVv6vCqciTM',
2617 'ext': 'mp4',
2618 'age_limit': 0,
2619 'comment_count': int,
2620 'chapters': 'count:13',
2621 'upload_date': '20221223',
2622 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2623 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2624 'like_count': int,
2625 'release_date': '20221223',
2626 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2627 '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】',
2628 'view_count': int,
2629 'playable_in_embed': True,
2630 'duration': 4438,
2631 'availability': 'public',
2632 'channel_follower_count': int,
2633 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2634 'categories': ['Entertainment'],
2635 'live_status': 'was_live',
2636 'release_timestamp': 1671793345,
2637 'channel': 'さなちゃんねる',
2638 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2639 'uploader': 'さなちゃんねる',
2640 'uploader_url': 'https://www.youtube.com/@sana_natori',
2641 'uploader_id': '@sana_natori',
2642 },
2643 },
2644 {
2645 # Fallbacks when webpage and web client is unavailable
2646 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2647 'info_dict': {
2648 'id': 'wSSmNUl9Snw',
2649 'ext': 'mp4',
2650 # 'categories': ['Science & Technology'],
2651 'view_count': int,
2652 'chapters': 'count:2',
2653 'channel': 'Scott Manley',
2654 'like_count': int,
2655 'age_limit': 0,
2656 # 'availability': 'public',
2657 'channel_follower_count': int,
2658 'live_status': 'not_live',
2659 'upload_date': '20170831',
2660 'duration': 682,
2661 'tags': 'count:8',
2662 'uploader_url': 'https://www.youtube.com/@scottmanley',
2663 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2664 'uploader': 'Scott Manley',
2665 'uploader_id': '@scottmanley',
2666 'title': 'The Computer Hack That Saved Apollo 14',
2667 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2668 'thumbnail': r're:^https?://.*\.webp',
2669 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2670 'playable_in_embed': True,
2671 },
2672 'params': {
2673 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2674 },
2675 },
2676 ]
2677
2678 _WEBPAGE_TESTS = [
2679 # YouTube <object> embed
2680 {
2681 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2682 'md5': '873c81d308b979f0e23ee7e620b312a3',
2683 'info_dict': {
2684 'id': 'msN87y-iEx0',
2685 'ext': 'mp4',
2686 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2687 'upload_date': '20080526',
2688 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2689 'age_limit': 0,
2690 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2691 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2692 'playable_in_embed': True,
2693 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2694 'like_count': int,
2695 'comment_count': int,
2696 'channel': 'Christopher Sykes',
2697 'live_status': 'not_live',
2698 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2699 'availability': 'public',
2700 'duration': 195,
2701 'view_count': int,
2702 'categories': ['Science & Technology'],
2703 'channel_follower_count': int,
2704 'uploader': 'Christopher Sykes',
2705 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2706 'uploader_id': '@ChristopherSykesDocumentaries',
2707 },
2708 'params': {
2709 'skip_download': True,
2710 }
2711 },
2712 ]
2713
2714 @classmethod
2715 def suitable(cls, url):
2716 from ..utils import parse_qs
2717
2718 qs = parse_qs(url)
2719 if qs.get('list', [None])[0]:
2720 return False
2721 return super().suitable(url)
2722
2723 def __init__(self, *args, **kwargs):
2724 super().__init__(*args, **kwargs)
2725 self._code_cache = {}
2726 self._player_cache = {}
2727
2728 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2729 lock = threading.Lock()
2730 start_time = time.time()
2731 formats = [f for f in formats if f.get('is_from_start')]
2732
2733 def refetch_manifest(format_id, delay):
2734 nonlocal formats, start_time, is_live
2735 if time.time() <= start_time + delay:
2736 return
2737
2738 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2739 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2740 microformats = traverse_obj(
2741 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2742 expected_type=dict)
2743 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2744 is_live = live_status == 'is_live'
2745 start_time = time.time()
2746
2747 def mpd_feed(format_id, delay):
2748 """
2749 @returns (manifest_url, manifest_stream_number, is_live) or None
2750 """
2751 for retry in self.RetryManager(fatal=False):
2752 with lock:
2753 refetch_manifest(format_id, delay)
2754
2755 f = next((f for f in formats if f['format_id'] == format_id), None)
2756 if not f:
2757 if not is_live:
2758 retry.error = f'{video_id}: Video is no longer live'
2759 else:
2760 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2761 continue
2762 return f['manifest_url'], f['manifest_stream_number'], is_live
2763 return None
2764
2765 for f in formats:
2766 f['is_live'] = is_live
2767 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2768 live_start_time, mpd_feed, not is_live and f.copy())
2769 if is_live:
2770 f['fragments'] = gen
2771 f['protocol'] = 'http_dash_segments_generator'
2772 else:
2773 f['fragments'] = LazyList(gen({}))
2774 del f['is_from_start']
2775
2776 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2777 FETCH_SPAN, MAX_DURATION = 5, 432000
2778
2779 mpd_url, stream_number, is_live = None, None, True
2780
2781 begin_index = 0
2782 download_start_time = ctx.get('start') or time.time()
2783
2784 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2785 if lack_early_segments:
2786 self.report_warning(bug_reports_message(
2787 'Starting download from the last 120 hours of the live stream since '
2788 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2789 lack_early_segments = True
2790
2791 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2792 fragments, fragment_base_url = None, None
2793
2794 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2795 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2796 # Obtain from MPD's maximum seq value
2797 old_mpd_url = mpd_url
2798 last_error = ctx.pop('last_error', None)
2799 expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2800 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2801 or (mpd_url, stream_number, False))
2802 if not refresh_sequence:
2803 if expire_fast and not is_live:
2804 return False, last_seq
2805 elif old_mpd_url == mpd_url:
2806 return True, last_seq
2807 if manifestless_orig_fmt:
2808 fmt_info = manifestless_orig_fmt
2809 else:
2810 try:
2811 fmts, _ = self._extract_mpd_formats_and_subtitles(
2812 mpd_url, None, note=False, errnote=False, fatal=False)
2813 except ExtractorError:
2814 fmts = None
2815 if not fmts:
2816 no_fragment_score += 2
2817 return False, last_seq
2818 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2819 fragments = fmt_info['fragments']
2820 fragment_base_url = fmt_info['fragment_base_url']
2821 assert fragment_base_url
2822
2823 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2824 return True, _last_seq
2825
2826 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2827 while is_live:
2828 fetch_time = time.time()
2829 if no_fragment_score > 30:
2830 return
2831 if last_segment_url:
2832 # Obtain from "X-Head-Seqnum" header value from each segment
2833 try:
2834 urlh = self._request_webpage(
2835 last_segment_url, None, note=False, errnote=False, fatal=False)
2836 except ExtractorError:
2837 urlh = None
2838 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2839 if last_seq is None:
2840 no_fragment_score += 2
2841 last_segment_url = None
2842 continue
2843 else:
2844 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2845 no_fragment_score += 2
2846 if not should_continue:
2847 continue
2848
2849 if known_idx > last_seq:
2850 last_segment_url = None
2851 continue
2852
2853 last_seq += 1
2854
2855 if begin_index < 0 and known_idx < 0:
2856 # skip from the start when it's negative value
2857 known_idx = last_seq + begin_index
2858 if lack_early_segments:
2859 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2860 try:
2861 for idx in range(known_idx, last_seq):
2862 # do not update sequence here or you'll get skipped some part of it
2863 should_continue, _ = _extract_sequence_from_mpd(False, False)
2864 if not should_continue:
2865 known_idx = idx - 1
2866 raise ExtractorError('breaking out of outer loop')
2867 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2868 yield {
2869 'url': last_segment_url,
2870 'fragment_count': last_seq,
2871 }
2872 if known_idx == last_seq:
2873 no_fragment_score += 5
2874 else:
2875 no_fragment_score = 0
2876 known_idx = last_seq
2877 except ExtractorError:
2878 continue
2879
2880 if manifestless_orig_fmt:
2881 # Stop at the first iteration if running for post-live manifestless;
2882 # fragment count no longer increase since it starts
2883 break
2884
2885 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2886
2887 def _extract_player_url(self, *ytcfgs, webpage=None):
2888 player_url = traverse_obj(
2889 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2890 get_all=False, expected_type=str)
2891 if not player_url:
2892 return
2893 return urljoin('https://www.youtube.com', player_url)
2894
2895 def _download_player_url(self, video_id, fatal=False):
2896 res = self._download_webpage(
2897 'https://www.youtube.com/iframe_api',
2898 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2899 if res:
2900 player_version = self._search_regex(
2901 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2902 if player_version:
2903 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2904
2905 def _signature_cache_id(self, example_sig):
2906 """ Return a string representation of a signature """
2907 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2908
2909 @classmethod
2910 def _extract_player_info(cls, player_url):
2911 for player_re in cls._PLAYER_INFO_RE:
2912 id_m = re.search(player_re, player_url)
2913 if id_m:
2914 break
2915 else:
2916 raise ExtractorError('Cannot identify player %r' % player_url)
2917 return id_m.group('id')
2918
2919 def _load_player(self, video_id, player_url, fatal=True):
2920 player_id = self._extract_player_info(player_url)
2921 if player_id not in self._code_cache:
2922 code = self._download_webpage(
2923 player_url, video_id, fatal=fatal,
2924 note='Downloading player ' + player_id,
2925 errnote='Download of %s failed' % player_url)
2926 if code:
2927 self._code_cache[player_id] = code
2928 return self._code_cache.get(player_id)
2929
2930 def _extract_signature_function(self, video_id, player_url, example_sig):
2931 player_id = self._extract_player_info(player_url)
2932
2933 # Read from filesystem cache
2934 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2935 assert os.path.basename(func_id) == func_id
2936
2937 self.write_debug(f'Extracting signature function {func_id}')
2938 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2939
2940 if not cache_spec:
2941 code = self._load_player(video_id, player_url)
2942 if code:
2943 res = self._parse_sig_js(code)
2944 test_string = ''.join(map(chr, range(len(example_sig))))
2945 cache_spec = [ord(c) for c in res(test_string)]
2946 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2947
2948 return lambda s: ''.join(s[i] for i in cache_spec)
2949
2950 def _print_sig_code(self, func, example_sig):
2951 if not self.get_param('youtube_print_sig_code'):
2952 return
2953
2954 def gen_sig_code(idxs):
2955 def _genslice(start, end, step):
2956 starts = '' if start == 0 else str(start)
2957 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2958 steps = '' if step == 1 else (':%d' % step)
2959 return f's[{starts}{ends}{steps}]'
2960
2961 step = None
2962 # Quelch pyflakes warnings - start will be set when step is set
2963 start = '(Never used)'
2964 for i, prev in zip(idxs[1:], idxs[:-1]):
2965 if step is not None:
2966 if i - prev == step:
2967 continue
2968 yield _genslice(start, prev, step)
2969 step = None
2970 continue
2971 if i - prev in [-1, 1]:
2972 step = i - prev
2973 start = prev
2974 continue
2975 else:
2976 yield 's[%d]' % prev
2977 if step is None:
2978 yield 's[%d]' % i
2979 else:
2980 yield _genslice(start, i, step)
2981
2982 test_string = ''.join(map(chr, range(len(example_sig))))
2983 cache_res = func(test_string)
2984 cache_spec = [ord(c) for c in cache_res]
2985 expr_code = ' + '.join(gen_sig_code(cache_spec))
2986 signature_id_tuple = '(%s)' % (
2987 ', '.join(str(len(p)) for p in example_sig.split('.')))
2988 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2989 ' return %s\n') % (signature_id_tuple, expr_code)
2990 self.to_screen('Extracted signature function:\n' + code)
2991
2992 def _parse_sig_js(self, jscode):
2993 funcname = self._search_regex(
2994 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2995 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2996 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2997 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2998 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)(?:;[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\))?',
2999 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3000 # Obsolete patterns
3001 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3002 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3003 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3004 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3005 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3006 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3007 jscode, 'Initial JS player signature function name', group='sig')
3008
3009 jsi = JSInterpreter(jscode)
3010 initial_function = jsi.extract_function(funcname)
3011 return lambda s: initial_function([s])
3012
3013 def _cached(self, func, *cache_id):
3014 def inner(*args, **kwargs):
3015 if cache_id not in self._player_cache:
3016 try:
3017 self._player_cache[cache_id] = func(*args, **kwargs)
3018 except ExtractorError as e:
3019 self._player_cache[cache_id] = e
3020 except Exception as e:
3021 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3022
3023 ret = self._player_cache[cache_id]
3024 if isinstance(ret, Exception):
3025 raise ret
3026 return ret
3027 return inner
3028
3029 def _decrypt_signature(self, s, video_id, player_url):
3030 """Turn the encrypted s field into a working signature"""
3031 extract_sig = self._cached(
3032 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3033 func = extract_sig(video_id, player_url, s)
3034 self._print_sig_code(func, s)
3035 return func(s)
3036
3037 def _decrypt_nsig(self, s, video_id, player_url):
3038 """Turn the encrypted n field into a working signature"""
3039 if player_url is None:
3040 raise ExtractorError('Cannot decrypt nsig without player_url')
3041 player_url = urljoin('https://www.youtube.com', player_url)
3042
3043 try:
3044 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3045 except ExtractorError as e:
3046 raise ExtractorError('Unable to extract nsig function code', cause=e)
3047 if self.get_param('youtube_print_sig_code'):
3048 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3049
3050 try:
3051 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3052 ret = extract_nsig(jsi, func_code)(s)
3053 except JSInterpreter.Exception as e:
3054 try:
3055 jsi = PhantomJSwrapper(self, timeout=5000)
3056 except ExtractorError:
3057 raise e
3058 self.report_warning(
3059 f'Native nsig extraction failed: Trying with PhantomJS\n'
3060 f' n = {s} ; player = {player_url}', video_id)
3061 self.write_debug(e, only_once=True)
3062
3063 args, func_body = func_code
3064 ret = jsi.execute(
3065 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3066 video_id=video_id, note='Executing signature code').strip()
3067
3068 self.write_debug(f'Decrypted nsig {s} => {ret}')
3069 return ret
3070
3071 def _extract_n_function_name(self, jscode):
3072 funcname, idx = self._search_regex(
3073 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3074 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3075 if not idx:
3076 return funcname
3077
3078 return json.loads(js_to_json(self._search_regex(
3079 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\]);', jscode,
3080 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3081
3082 def _extract_n_function_code(self, video_id, player_url):
3083 player_id = self._extract_player_info(player_url)
3084 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3085 jscode = func_code or self._load_player(video_id, player_url)
3086 jsi = JSInterpreter(jscode)
3087
3088 if func_code:
3089 return jsi, player_id, func_code
3090
3091 func_name = self._extract_n_function_name(jscode)
3092
3093 # For redundancy
3094 func_code = self._search_regex(
3095 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3096 # NB: The end of the regex is intentionally kept strict
3097 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3098 jscode, 'nsig function', group=('var', 'code'), default=None)
3099 if func_code:
3100 func_code = ([func_code[0]], func_code[1])
3101 else:
3102 self.write_debug('Extracting nsig function with jsinterp')
3103 func_code = jsi.extract_function_code(func_name)
3104
3105 self.cache.store('youtube-nsig', player_id, func_code)
3106 return jsi, player_id, func_code
3107
3108 def _extract_n_function_from_code(self, jsi, func_code):
3109 func = jsi.extract_function_from_code(*func_code)
3110
3111 def extract_nsig(s):
3112 try:
3113 ret = func([s])
3114 except JSInterpreter.Exception:
3115 raise
3116 except Exception as e:
3117 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3118
3119 if ret.startswith('enhanced_except_'):
3120 raise JSInterpreter.Exception('Signature function returned an exception')
3121 return ret
3122
3123 return extract_nsig
3124
3125 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3126 """
3127 Extract signatureTimestamp (sts)
3128 Required to tell API what sig/player version is in use.
3129 """
3130 sts = None
3131 if isinstance(ytcfg, dict):
3132 sts = int_or_none(ytcfg.get('STS'))
3133
3134 if not sts:
3135 # Attempt to extract from player
3136 if player_url is None:
3137 error_msg = 'Cannot extract signature timestamp without player_url.'
3138 if fatal:
3139 raise ExtractorError(error_msg)
3140 self.report_warning(error_msg)
3141 return
3142 code = self._load_player(video_id, player_url, fatal=fatal)
3143 if code:
3144 sts = int_or_none(self._search_regex(
3145 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3146 'JS player signature timestamp', group='sts', fatal=fatal))
3147 return sts
3148
3149 def _mark_watched(self, video_id, player_responses):
3150 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3151 label = 'fully ' if is_full else ''
3152 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3153 expected_type=url_or_none)
3154 if not url:
3155 self.report_warning(f'Unable to mark {label}watched')
3156 return
3157 parsed_url = urllib.parse.urlparse(url)
3158 qs = urllib.parse.parse_qs(parsed_url.query)
3159
3160 # cpn generation algorithm is reverse engineered from base.js.
3161 # In fact it works even with dummy cpn.
3162 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3163 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3164
3165 # # more consistent results setting it to right before the end
3166 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3167
3168 qs.update({
3169 'ver': ['2'],
3170 'cpn': [cpn],
3171 'cmt': video_length,
3172 'el': 'detailpage', # otherwise defaults to "shorts"
3173 })
3174
3175 if is_full:
3176 # these seem to mark watchtime "history" in the real world
3177 # they're required, so send in a single value
3178 qs.update({
3179 'st': 0,
3180 'et': video_length,
3181 })
3182
3183 url = urllib.parse.urlunparse(
3184 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3185
3186 self._download_webpage(
3187 url, video_id, f'Marking {label}watched',
3188 'Unable to mark watched', fatal=False)
3189
3190 @classmethod
3191 def _extract_from_webpage(cls, url, webpage):
3192 # Invidious Instances
3193 # https://github.com/yt-dlp/yt-dlp/issues/195
3194 # https://github.com/iv-org/invidious/pull/1730
3195 mobj = re.search(
3196 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3197 webpage)
3198 if mobj:
3199 yield cls.url_result(mobj.group('url'), cls)
3200 raise cls.StopExtraction()
3201
3202 yield from super()._extract_from_webpage(url, webpage)
3203
3204 # lazyYT YouTube embed
3205 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3206 yield cls.url_result(unescapeHTML(id_), cls, id_)
3207
3208 # Wordpress "YouTube Video Importer" plugin
3209 for m in re.findall(r'''(?x)<div[^>]+
3210 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3211 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3212 yield cls.url_result(m[-1], cls, m[-1])
3213
3214 @classmethod
3215 def extract_id(cls, url):
3216 video_id = cls.get_temp_id(url)
3217 if not video_id:
3218 raise ExtractorError(f'Invalid URL: {url}')
3219 return video_id
3220
3221 def _extract_chapters_from_json(self, data, duration):
3222 chapter_list = traverse_obj(
3223 data, (
3224 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3225 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3226 ), expected_type=list)
3227
3228 return self._extract_chapters_helper(
3229 chapter_list,
3230 start_function=lambda chapter: float_or_none(
3231 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3232 title_function=lambda chapter: traverse_obj(
3233 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3234 duration=duration)
3235
3236 def _extract_chapters_from_engagement_panel(self, data, duration):
3237 content_list = traverse_obj(
3238 data,
3239 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3240 expected_type=list)
3241 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3242 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3243
3244 return next(filter(None, (
3245 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3246 chapter_time, chapter_title, duration)
3247 for contents in content_list)), [])
3248
3249 def _extract_heatmap_from_player_overlay(self, data):
3250 content_list = traverse_obj(data, (
3251 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer', 'decoratedPlayerBarRenderer', 'playerBar',
3252 'multiMarkersPlayerBarRenderer', 'markersMap', ..., 'value', 'heatmap', 'heatmapRenderer', 'heatMarkers', {list}))
3253 return next(filter(None, (
3254 traverse_obj(contents, (..., 'heatMarkerRenderer', {
3255 'start_time': ('timeRangeStartMillis', {functools.partial(float_or_none, scale=1000)}),
3256 'end_time': {lambda x: (x['timeRangeStartMillis'] + x['markerDurationMillis']) / 1000},
3257 'value': ('heatMarkerIntensityScoreNormalized', {float_or_none}),
3258 })) for contents in content_list)), None)
3259
3260 def _extract_comment(self, comment_renderer, parent=None):
3261 comment_id = comment_renderer.get('commentId')
3262 if not comment_id:
3263 return
3264
3265 text = self._get_text(comment_renderer, 'contentText')
3266
3267 # Timestamp is an estimate calculated from the current time and time_text
3268 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3269 timestamp = self._parse_time_text(time_text)
3270
3271 author = self._get_text(comment_renderer, 'authorText')
3272 author_id = try_get(comment_renderer,
3273 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], str)
3274
3275 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
3276 lambda x: x['likeCount']), str)) or 0
3277 author_thumbnail = try_get(comment_renderer,
3278 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], str)
3279
3280 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
3281 is_favorited = 'creatorHeart' in (try_get(
3282 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
3283 return {
3284 'id': comment_id,
3285 'text': text,
3286 'timestamp': timestamp,
3287 'time_text': time_text,
3288 'like_count': votes,
3289 'is_favorited': is_favorited,
3290 'author': author,
3291 'author_id': author_id,
3292 'author_thumbnail': author_thumbnail,
3293 'author_is_uploader': author_is_uploader,
3294 'parent': parent or 'root'
3295 }
3296
3297 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3298
3299 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3300
3301 def extract_header(contents):
3302 _continuation = None
3303 for content in contents:
3304 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3305 expected_comment_count = self._get_count(
3306 comments_header_renderer, 'countText', 'commentsCount')
3307
3308 if expected_comment_count:
3309 tracker['est_total'] = expected_comment_count
3310 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3311 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3312
3313 sort_menu_item = try_get(
3314 comments_header_renderer,
3315 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3316 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3317
3318 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3319 if not _continuation:
3320 continue
3321
3322 sort_text = str_or_none(sort_menu_item.get('title'))
3323 if not sort_text:
3324 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3325 self.to_screen('Sorting comments by %s' % sort_text.lower())
3326 break
3327 return _continuation
3328
3329 def extract_thread(contents):
3330 if not parent:
3331 tracker['current_page_thread'] = 0
3332 for content in contents:
3333 if not parent and tracker['total_parent_comments'] >= max_parents:
3334 yield
3335 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3336 comment_renderer = get_first(
3337 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3338 expected_type=dict, default={})
3339
3340 comment = self._extract_comment(comment_renderer, parent)
3341 if not comment:
3342 continue
3343 is_pinned = bool(traverse_obj(comment_renderer, 'pinnedCommentBadge'))
3344 comment_id = comment['id']
3345 if is_pinned:
3346 tracker['pinned_comment_ids'].add(comment_id)
3347 # Sometimes YouTube may break and give us infinite looping comments.
3348 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3349 if comment_id in tracker['seen_comment_ids']:
3350 if comment_id in tracker['pinned_comment_ids'] and not is_pinned:
3351 # Pinned comments may appear a second time in newest first sort
3352 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3353 continue
3354 self.report_warning('Detected YouTube comments looping. Stopping comment extraction as we probably cannot get any more.')
3355 yield
3356 else:
3357 tracker['seen_comment_ids'].add(comment['id'])
3358
3359 tracker['running_total'] += 1
3360 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3361 yield comment
3362
3363 # Attempt to get the replies
3364 comment_replies_renderer = try_get(
3365 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3366
3367 if comment_replies_renderer:
3368 tracker['current_page_thread'] += 1
3369 comment_entries_iter = self._comment_entries(
3370 comment_replies_renderer, ytcfg, video_id,
3371 parent=comment.get('id'), tracker=tracker)
3372 yield from itertools.islice(comment_entries_iter, min(
3373 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3374
3375 # Keeps track of counts across recursive calls
3376 if not tracker:
3377 tracker = dict(
3378 running_total=0,
3379 est_total=0,
3380 current_page_thread=0,
3381 total_parent_comments=0,
3382 total_reply_comments=0,
3383 seen_comment_ids=set(),
3384 pinned_comment_ids=set()
3385 )
3386
3387 # TODO: Deprecated
3388 # YouTube comments have a max depth of 2
3389 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3390 if max_depth:
3391 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3392 'Set max replies in the max-comments extractor argument instead')
3393 if max_depth == 1 and parent:
3394 return
3395
3396 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3397 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3398
3399 continuation = self._extract_continuation(root_continuation_data)
3400
3401 response = None
3402 is_forced_continuation = False
3403 is_first_continuation = parent is None
3404 if is_first_continuation and not continuation:
3405 # Sometimes you can get comments by generating the continuation yourself,
3406 # even if YouTube initially reports them being disabled - e.g. stories comments.
3407 # Note: if the comment section is actually disabled, YouTube may return a response with
3408 # required check_get_keys missing. So we will disable that check initially in this case.
3409 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3410 is_forced_continuation = True
3411
3412 for page_num in itertools.count(0):
3413 if not continuation:
3414 break
3415 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3416 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
3417 if page_num == 0:
3418 if is_first_continuation:
3419 note_prefix = 'Downloading comment section API JSON'
3420 else:
3421 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3422 tracker['current_page_thread'], comment_prog_str)
3423 else:
3424 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3425 ' ' if parent else '', ' replies' if parent else '',
3426 page_num, comment_prog_str)
3427 try:
3428 response = self._extract_response(
3429 item_id=None, query=continuation,
3430 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3431 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
3432 except ExtractorError as e:
3433 # Ignore incomplete data error for replies if retries didn't work.
3434 # This is to allow any other parent comments and comment threads to be downloaded.
3435 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3436 if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
3437 self.report_warning(
3438 'Received incomplete data for a comment reply thread and retrying did not help. '
3439 'Ignoring to let other comments be downloaded.')
3440 else:
3441 raise
3442 is_forced_continuation = False
3443 continuation_contents = traverse_obj(
3444 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
3445
3446 continuation = None
3447 for continuation_section in continuation_contents:
3448 continuation_items = traverse_obj(
3449 continuation_section,
3450 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
3451 get_all=False, expected_type=list) or []
3452 if is_first_continuation:
3453 continuation = extract_header(continuation_items)
3454 is_first_continuation = False
3455 if continuation:
3456 break
3457 continue
3458
3459 for entry in extract_thread(continuation_items):
3460 if not entry:
3461 return
3462 yield entry
3463 continuation = self._extract_continuation({'contents': continuation_items})
3464 if continuation:
3465 break
3466
3467 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3468 if message and not parent and tracker['running_total'] == 0:
3469 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3470 raise self.CommentsDisabled
3471
3472 @staticmethod
3473 def _generate_comment_continuation(video_id):
3474 """
3475 Generates initial comment section continuation token from given video id
3476 """
3477 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3478 return base64.b64encode(token.encode()).decode()
3479
3480 def _get_comments(self, ytcfg, video_id, contents, webpage):
3481 """Entry for comment extraction"""
3482 def _real_comment_extract(contents):
3483 renderer = next((
3484 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3485 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3486 yield from self._comment_entries(renderer, ytcfg, video_id)
3487
3488 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3489 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3490
3491 @staticmethod
3492 def _get_checkok_params():
3493 return {'contentCheckOk': True, 'racyCheckOk': True}
3494
3495 @classmethod
3496 def _generate_player_context(cls, sts=None):
3497 context = {
3498 'html5Preference': 'HTML5_PREF_WANTS',
3499 }
3500 if sts is not None:
3501 context['signatureTimestamp'] = sts
3502 return {
3503 'playbackContext': {
3504 'contentPlaybackContext': context
3505 },
3506 **cls._get_checkok_params()
3507 }
3508
3509 @staticmethod
3510 def _is_agegated(player_response):
3511 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3512 return True
3513
3514 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3515 AGE_GATE_REASONS = (
3516 'confirm your age', 'age-restricted', 'inappropriate', # reason
3517 'age_verification_required', 'age_check_required', # status
3518 )
3519 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3520
3521 @staticmethod
3522 def _is_unplayable(player_response):
3523 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3524
3525 _STORY_PLAYER_PARAMS = '8AEB'
3526
3527 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3528
3529 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3530 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3531 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3532 headers = self.generate_api_headers(
3533 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3534
3535 yt_query = {
3536 'videoId': video_id,
3537 }
3538 if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
3539 yt_query['params'] = self._STORY_PLAYER_PARAMS
3540
3541 yt_query.update(self._generate_player_context(sts))
3542 return self._extract_response(
3543 item_id=video_id, ep='player', query=yt_query,
3544 ytcfg=player_ytcfg, headers=headers, fatal=True,
3545 default_client=client,
3546 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3547 ) or None
3548
3549 def _get_requested_clients(self, url, smuggled_data):
3550 requested_clients = []
3551 default = ['android', 'web']
3552 allowed_clients = sorted(
3553 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3554 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3555 for client in self._configuration_arg('player_client'):
3556 if client in allowed_clients:
3557 requested_clients.append(client)
3558 elif client == 'default':
3559 requested_clients.extend(default)
3560 elif client == 'all':
3561 requested_clients.extend(allowed_clients)
3562 else:
3563 self.report_warning(f'Skipping unsupported client {client}')
3564 if not requested_clients:
3565 requested_clients = default
3566
3567 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3568 requested_clients.extend(
3569 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3570
3571 return orderedSet(requested_clients)
3572
3573 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3574 initial_pr = None
3575 if webpage:
3576 initial_pr = self._search_json(
3577 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3578
3579 all_clients = set(clients)
3580 clients = clients[::-1]
3581 prs = []
3582
3583 def append_client(*client_names):
3584 """ Append the first client name that exists but not already used """
3585 for client_name in client_names:
3586 actual_client = _split_innertube_client(client_name)[0]
3587 if actual_client in INNERTUBE_CLIENTS:
3588 if actual_client not in all_clients:
3589 clients.append(client_name)
3590 all_clients.add(actual_client)
3591 return
3592
3593 # Android player_response does not have microFormats which are needed for
3594 # extraction of some data. So we return the initial_pr with formats
3595 # stripped out even if not requested by the user
3596 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3597 if initial_pr:
3598 pr = dict(initial_pr)
3599 pr['streamingData'] = None
3600 prs.append(pr)
3601
3602 last_error = None
3603 tried_iframe_fallback = False
3604 player_url = None
3605 while clients:
3606 client, base_client, variant = _split_innertube_client(clients.pop())
3607 player_ytcfg = master_ytcfg if client == 'web' else {}
3608 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3609 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3610
3611 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3612 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3613 if 'js' in self._configuration_arg('player_skip'):
3614 require_js_player = False
3615 player_url = None
3616
3617 if not player_url and not tried_iframe_fallback and require_js_player:
3618 player_url = self._download_player_url(video_id)
3619 tried_iframe_fallback = True
3620
3621 try:
3622 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3623 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3624 except ExtractorError as e:
3625 if last_error:
3626 self.report_warning(last_error)
3627 last_error = e
3628 continue
3629
3630 if pr:
3631 # YouTube may return a different video player response than expected.
3632 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3633 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3634 if pr_video_id and pr_video_id != video_id:
3635 self.report_warning(
3636 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3637 else:
3638 # Save client name for introspection later
3639 name = short_client_name(client)
3640 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3641 sd[STREAMING_DATA_CLIENT_NAME] = name
3642 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3643 f[STREAMING_DATA_CLIENT_NAME] = name
3644 prs.append(pr)
3645
3646 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3647 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3648 append_client(f'{base_client}_creator')
3649 elif self._is_agegated(pr):
3650 if variant == 'tv_embedded':
3651 append_client(f'{base_client}_embedded')
3652 elif not variant:
3653 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3654
3655 if last_error:
3656 if not len(prs):
3657 raise last_error
3658 self.report_warning(last_error)
3659 return prs, player_url
3660
3661 def _needs_live_processing(self, live_status, duration):
3662 if (live_status == 'is_live' and self.get_param('live_from_start')
3663 or live_status == 'post_live' and (duration or 0) > 4 * 3600):
3664 return live_status
3665
3666 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3667 CHUNK_SIZE = 10 << 20
3668 itags, stream_ids = collections.defaultdict(set), []
3669 itag_qualities, res_qualities = {}, {0: None}
3670 q = qualities([
3671 # Normally tiny is the smallest video-only formats. But
3672 # audio-only formats with unknown quality may get tagged as tiny
3673 'tiny',
3674 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3675 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3676 ])
3677 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3678 all_formats = self._configuration_arg('include_duplicate_formats')
3679
3680 def build_fragments(f):
3681 return LazyList({
3682 'url': update_url_query(f['url'], {
3683 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3684 })
3685 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3686
3687 for fmt in streaming_formats:
3688 if fmt.get('targetDurationSec'):
3689 continue
3690
3691 itag = str_or_none(fmt.get('itag'))
3692 audio_track = fmt.get('audioTrack') or {}
3693 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3694 if not all_formats:
3695 if stream_id in stream_ids:
3696 continue
3697
3698 quality = fmt.get('quality')
3699 height = int_or_none(fmt.get('height'))
3700 if quality == 'tiny' or not quality:
3701 quality = fmt.get('audioQuality', '').lower() or quality
3702 # The 3gp format (17) in android client has a quality of "small",
3703 # but is actually worse than other formats
3704 if itag == '17':
3705 quality = 'tiny'
3706 if quality:
3707 if itag:
3708 itag_qualities[itag] = quality
3709 if height:
3710 res_qualities[height] = quality
3711 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3712 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3713 # number of fragment that would subsequently requested with (`&sq=N`)
3714 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3715 continue
3716
3717 fmt_url = fmt.get('url')
3718 if not fmt_url:
3719 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3720 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3721 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3722 if not all((sc, fmt_url, player_url, encrypted_sig)):
3723 continue
3724 try:
3725 fmt_url += '&%s=%s' % (
3726 traverse_obj(sc, ('sp', -1)) or 'signature',
3727 self._decrypt_signature(encrypted_sig, video_id, player_url)
3728 )
3729 except ExtractorError as e:
3730 self.report_warning('Signature extraction failed: Some formats may be missing',
3731 video_id=video_id, only_once=True)
3732 self.write_debug(e, only_once=True)
3733 continue
3734
3735 query = parse_qs(fmt_url)
3736 throttled = False
3737 if query.get('n'):
3738 try:
3739 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3740 fmt_url = update_url_query(fmt_url, {
3741 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3742 })
3743 except ExtractorError as e:
3744 phantomjs_hint = ''
3745 if isinstance(e, JSInterpreter.Exception):
3746 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3747 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3748 if player_url:
3749 self.report_warning(
3750 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3751 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3752 self.write_debug(e, only_once=True)
3753 else:
3754 self.report_warning(
3755 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3756 video_id=video_id, only_once=True)
3757 throttled = True
3758
3759 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3760 language_preference = (
3761 10 if audio_track.get('audioIsDefault') and 10
3762 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3763 else -1)
3764 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3765 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3766 # Make sure to avoid false positives with small duration differences.
3767 # E.g. __2ABJjxzNo, ySuUZEjARPY
3768 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3769 if is_damaged:
3770 self.report_warning(
3771 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3772
3773 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3774 dct = {
3775 'asr': int_or_none(fmt.get('audioSampleRate')),
3776 'filesize': int_or_none(fmt.get('contentLength')),
3777 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3778 'format_note': join_nonempty(
3779 join_nonempty(audio_track.get('displayName'),
3780 language_preference > 0 and ' (default)', delim=''),
3781 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3782 fmt.get('isDrc') and 'DRC',
3783 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3784 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3785 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3786 (self.get_param('verbose') or all_formats) and client_name,
3787 delim=', '),
3788 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3789 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3790 'fps': int_or_none(fmt.get('fps')) or None,
3791 'audio_channels': fmt.get('audioChannels'),
3792 'height': height,
3793 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3794 'has_drm': bool(fmt.get('drmFamilies')),
3795 'tbr': tbr,
3796 'url': fmt_url,
3797 'width': int_or_none(fmt.get('width')),
3798 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3799 'desc' if language_preference < -1 else '') or None,
3800 'language_preference': language_preference,
3801 # Strictly de-prioritize damaged and 3gp formats
3802 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3803 }
3804 mime_mobj = re.match(
3805 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3806 if mime_mobj:
3807 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3808 dct.update(parse_codecs(mime_mobj.group(2)))
3809 if itag:
3810 itags[itag].add(('https', dct.get('language')))
3811 stream_ids.append(stream_id)
3812 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3813 if single_stream and dct.get('ext'):
3814 dct['container'] = dct['ext'] + '_dash'
3815
3816 if all_formats and dct['filesize']:
3817 yield {
3818 **dct,
3819 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3820 'protocol': 'http_dash_segments',
3821 'fragments': build_fragments(dct),
3822 }
3823 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3824 yield dct
3825
3826 needs_live_processing = self._needs_live_processing(live_status, duration)
3827 skip_bad_formats = not self._configuration_arg('include_incomplete_formats')
3828
3829 skip_manifests = set(self._configuration_arg('skip'))
3830 if (not self.get_param('youtube_include_hls_manifest', True)
3831 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3832 or needs_live_processing and skip_bad_formats):
3833 skip_manifests.add('hls')
3834
3835 if not self.get_param('youtube_include_dash_manifest', True):
3836 skip_manifests.add('dash')
3837 if self._configuration_arg('include_live_dash'):
3838 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3839 'Use include_incomplete_formats extractor argument instead')
3840 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3841 skip_manifests.add('dash')
3842
3843 def process_manifest_format(f, proto, client_name, itag):
3844 key = (proto, f.get('language'))
3845 if not all_formats and key in itags[itag]:
3846 return False
3847 itags[itag].add(key)
3848
3849 if itag and all_formats:
3850 f['format_id'] = f'{itag}-{proto}'
3851 elif any(p != proto for p, _ in itags[itag]):
3852 f['format_id'] = f'{itag}-{proto}'
3853 elif itag:
3854 f['format_id'] = itag
3855
3856 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3857 if f['quality'] == -1 and f.get('height'):
3858 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3859 if self.get_param('verbose'):
3860 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3861 return True
3862
3863 subtitles = {}
3864 for sd in streaming_data:
3865 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3866
3867 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3868 if hls_manifest_url:
3869 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3870 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3871 subtitles = self._merge_subtitles(subs, subtitles)
3872 for f in fmts:
3873 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3874 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3875 yield f
3876
3877 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3878 if dash_manifest_url:
3879 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3880 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3881 for f in formats:
3882 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3883 f['filesize'] = int_or_none(self._search_regex(
3884 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3885 if needs_live_processing:
3886 f['is_from_start'] = True
3887
3888 yield f
3889 yield subtitles
3890
3891 def _extract_storyboard(self, player_responses, duration):
3892 spec = get_first(
3893 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3894 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3895 if not base_url:
3896 return
3897 L = len(spec) - 1
3898 for i, args in enumerate(spec):
3899 args = args.split('#')
3900 counts = list(map(int_or_none, args[:5]))
3901 if len(args) != 8 or not all(counts):
3902 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3903 continue
3904 width, height, frame_count, cols, rows = counts
3905 N, sigh = args[6:]
3906
3907 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3908 fragment_count = frame_count / (cols * rows)
3909 fragment_duration = duration / fragment_count
3910 yield {
3911 'format_id': f'sb{i}',
3912 'format_note': 'storyboard',
3913 'ext': 'mhtml',
3914 'protocol': 'mhtml',
3915 'acodec': 'none',
3916 'vcodec': 'none',
3917 'url': url,
3918 'width': width,
3919 'height': height,
3920 'fps': frame_count / duration,
3921 'rows': rows,
3922 'columns': cols,
3923 'fragments': [{
3924 'url': url.replace('$M', str(j)),
3925 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3926 } for j in range(math.ceil(fragment_count))],
3927 }
3928
3929 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3930 webpage = None
3931 if 'webpage' not in self._configuration_arg('player_skip'):
3932 query = {'bpctr': '9999999999', 'has_verified': '1'}
3933 if smuggled_data.get('is_story'):
3934 query['pp'] = self._STORY_PLAYER_PARAMS
3935 webpage = self._download_webpage(
3936 webpage_url, video_id, fatal=False, query=query)
3937
3938 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3939
3940 player_responses, player_url = self._extract_player_responses(
3941 self._get_requested_clients(url, smuggled_data),
3942 video_id, webpage, master_ytcfg, smuggled_data)
3943
3944 return webpage, master_ytcfg, player_responses, player_url
3945
3946 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3947 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3948 is_live = get_first(video_details, 'isLive')
3949 if is_live is None:
3950 is_live = get_first(live_broadcast_details, 'isLiveNow')
3951 live_content = get_first(video_details, 'isLiveContent')
3952 is_upcoming = get_first(video_details, 'isUpcoming')
3953 post_live = get_first(video_details, 'isPostLiveDvr')
3954 live_status = ('post_live' if post_live
3955 else 'is_live' if is_live
3956 else 'is_upcoming' if is_upcoming
3957 else 'was_live' if live_content
3958 else 'not_live' if False in (is_live, live_content)
3959 else None)
3960 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
3961 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
3962
3963 return live_broadcast_details, live_status, streaming_data, formats, subtitles
3964
3965 def _real_extract(self, url):
3966 url, smuggled_data = unsmuggle_url(url, {})
3967 video_id = self._match_id(url)
3968
3969 base_url = self.http_scheme() + '//www.youtube.com/'
3970 webpage_url = base_url + 'watch?v=' + video_id
3971
3972 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3973
3974 playability_statuses = traverse_obj(
3975 player_responses, (..., 'playabilityStatus'), expected_type=dict)
3976
3977 trailer_video_id = get_first(
3978 playability_statuses,
3979 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3980 expected_type=str)
3981 if trailer_video_id:
3982 return self.url_result(
3983 trailer_video_id, self.ie_key(), trailer_video_id)
3984
3985 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3986 if webpage else (lambda x: None))
3987
3988 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
3989 microformats = traverse_obj(
3990 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3991 expected_type=dict)
3992
3993 translated_title = self._get_text(microformats, (..., 'title'))
3994 video_title = (self._preferred_lang and translated_title
3995 or get_first(video_details, 'title') # primary
3996 or translated_title
3997 or search_meta(['og:title', 'twitter:title', 'title']))
3998 translated_description = self._get_text(microformats, (..., 'description'))
3999 original_description = get_first(video_details, 'shortDescription')
4000 video_description = (
4001 self._preferred_lang and translated_description
4002 # If original description is blank, it will be an empty string.
4003 # Do not prefer translated description in this case.
4004 or original_description if original_description is not None else translated_description)
4005
4006 multifeed_metadata_list = get_first(
4007 player_responses,
4008 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4009 expected_type=str)
4010 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4011 if self.get_param('noplaylist'):
4012 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4013 else:
4014 entries = []
4015 feed_ids = []
4016 for feed in multifeed_metadata_list.split(','):
4017 # Unquote should take place before split on comma (,) since textual
4018 # fields may contain comma as well (see
4019 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4020 feed_data = urllib.parse.parse_qs(
4021 urllib.parse.unquote_plus(feed))
4022
4023 def feed_entry(name):
4024 return try_get(
4025 feed_data, lambda x: x[name][0], str)
4026
4027 feed_id = feed_entry('id')
4028 if not feed_id:
4029 continue
4030 feed_title = feed_entry('title')
4031 title = video_title
4032 if feed_title:
4033 title += ' (%s)' % feed_title
4034 entries.append({
4035 '_type': 'url_transparent',
4036 'ie_key': 'Youtube',
4037 'url': smuggle_url(
4038 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4039 {'force_singlefeed': True}),
4040 'title': title,
4041 })
4042 feed_ids.append(feed_id)
4043 self.to_screen(
4044 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4045 % (', '.join(feed_ids), video_id))
4046 return self.playlist_result(
4047 entries, video_id, video_title, video_description)
4048
4049 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4050 or int_or_none(get_first(microformats, 'lengthSeconds'))
4051 or parse_duration(search_meta('duration')) or None)
4052
4053 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4054 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4055 if live_status == 'post_live':
4056 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4057
4058 if not formats:
4059 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4060 self.report_drm(video_id)
4061 pemr = get_first(
4062 playability_statuses,
4063 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4064 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4065 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4066 if subreason:
4067 if subreason == 'The uploader has not made this video available in your country.':
4068 countries = get_first(microformats, 'availableCountries')
4069 if not countries:
4070 regions_allowed = search_meta('regionsAllowed')
4071 countries = regions_allowed.split(',') if regions_allowed else None
4072 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4073 reason += f'. {subreason}'
4074 if reason:
4075 self.raise_no_formats(reason, expected=True)
4076
4077 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4078 if not keywords and webpage:
4079 keywords = [
4080 unescapeHTML(m.group('content'))
4081 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4082 for keyword in keywords:
4083 if keyword.startswith('yt:stretch='):
4084 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4085 if mobj:
4086 # NB: float is intentional for forcing float division
4087 w, h = (float(v) for v in mobj.groups())
4088 if w > 0 and h > 0:
4089 ratio = w / h
4090 for f in formats:
4091 if f.get('vcodec') != 'none':
4092 f['stretched_ratio'] = ratio
4093 break
4094 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4095 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4096 if thumbnail_url:
4097 thumbnails.append({
4098 'url': thumbnail_url,
4099 })
4100 original_thumbnails = thumbnails.copy()
4101
4102 # The best resolution thumbnails sometimes does not appear in the webpage
4103 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4104 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4105 thumbnail_names = [
4106 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4107 # in resolution, these are not the custom thumbnail. So de-prioritize them
4108 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4109 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4110 ]
4111 n_thumbnail_names = len(thumbnail_names)
4112 thumbnails.extend({
4113 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4114 video_id=video_id, name=name, ext=ext,
4115 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4116 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4117 for thumb in thumbnails:
4118 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4119 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4120 self._remove_duplicate_formats(thumbnails)
4121 self._downloader._sort_thumbnails(original_thumbnails)
4122
4123 category = get_first(microformats, 'category') or search_meta('genre')
4124 channel_id = self.ucid_or_none(str_or_none(
4125 get_first(video_details, 'channelId')
4126 or get_first(microformats, 'externalChannelId')
4127 or search_meta('channelId')))
4128 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4129
4130 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4131 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4132 if not duration and live_end_time and live_start_time:
4133 duration = live_end_time - live_start_time
4134
4135 needs_live_processing = self._needs_live_processing(live_status, duration)
4136
4137 def is_bad_format(fmt):
4138 if needs_live_processing and not fmt.get('is_from_start'):
4139 return True
4140 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4141 and fmt.get('protocol') == 'http_dash_segments'):
4142 return True
4143
4144 for fmt in filter(is_bad_format, formats):
4145 fmt['preference'] = (fmt.get('preference') or -1) - 10
4146 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 4 hours)', delim=' ')
4147
4148 if needs_live_processing:
4149 self._prepare_live_from_start_formats(
4150 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4151
4152 formats.extend(self._extract_storyboard(player_responses, duration))
4153
4154 channel_handle = self.handle_from_url(owner_profile_url)
4155
4156 info = {
4157 'id': video_id,
4158 'title': video_title,
4159 'formats': formats,
4160 'thumbnails': thumbnails,
4161 # The best thumbnail that we are sure exists. Prevents unnecessary
4162 # URL checking if user don't care about getting the best possible thumbnail
4163 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4164 'description': video_description,
4165 'channel_id': channel_id,
4166 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4167 'duration': duration,
4168 'view_count': int_or_none(
4169 get_first((video_details, microformats), (..., 'viewCount'))
4170 or search_meta('interactionCount')),
4171 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4172 'age_limit': 18 if (
4173 get_first(microformats, 'isFamilySafe') is False
4174 or search_meta('isFamilyFriendly') == 'false'
4175 or search_meta('og:restrictions:age') == '18+') else 0,
4176 'webpage_url': webpage_url,
4177 'categories': [category] if category else None,
4178 'tags': keywords,
4179 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4180 'live_status': live_status,
4181 'release_timestamp': live_start_time,
4182 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4183 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4184 }
4185
4186 subtitles = {}
4187 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4188 if pctr:
4189 def get_lang_code(track):
4190 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4191 or track.get('languageCode'))
4192
4193 # Converted into dicts to remove duplicates
4194 captions = {
4195 get_lang_code(sub): sub
4196 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4197 translation_languages = {
4198 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4199 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4200
4201 def process_language(container, base_url, lang_code, sub_name, query):
4202 lang_subs = container.setdefault(lang_code, [])
4203 for fmt in self._SUBTITLE_FORMATS:
4204 query.update({
4205 'fmt': fmt,
4206 })
4207 lang_subs.append({
4208 'ext': fmt,
4209 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4210 'name': sub_name,
4211 })
4212
4213 # NB: Constructing the full subtitle dictionary is slow
4214 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4215 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4216 for lang_code, caption_track in captions.items():
4217 base_url = caption_track.get('baseUrl')
4218 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4219 if not base_url:
4220 continue
4221 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4222 if caption_track.get('kind') != 'asr':
4223 if not lang_code:
4224 continue
4225 process_language(
4226 subtitles, base_url, lang_code, lang_name, {})
4227 if not caption_track.get('isTranslatable'):
4228 continue
4229 for trans_code, trans_name in translation_languages.items():
4230 if not trans_code:
4231 continue
4232 orig_trans_code = trans_code
4233 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4234 if not get_translated_subs:
4235 continue
4236 trans_code += f'-{lang_code}'
4237 trans_name += format_field(lang_name, None, ' from %s')
4238 # Add an "-orig" label to the original language so that it can be distinguished.
4239 # The subs are returned without "-orig" as well for compatibility
4240 if lang_code == f'a-{orig_trans_code}':
4241 process_language(
4242 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4243 # Setting tlang=lang returns damaged subtitles.
4244 process_language(automatic_captions, base_url, trans_code, trans_name,
4245 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4246
4247 info['automatic_captions'] = automatic_captions
4248 info['subtitles'] = subtitles
4249
4250 parsed_url = urllib.parse.urlparse(url)
4251 for component in [parsed_url.fragment, parsed_url.query]:
4252 query = urllib.parse.parse_qs(component)
4253 for k, v in query.items():
4254 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4255 d_k += '_time'
4256 if d_k not in info and k in s_ks:
4257 info[d_k] = parse_duration(query[k][0])
4258
4259 # Youtube Music Auto-generated description
4260 if video_description:
4261 mobj = re.search(
4262 r'''(?xs)
4263 (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
4264 (?P<album>[^\n]+)
4265 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4266 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4267 (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
4268 .+\nAuto-generated\ by\ YouTube\.\s*$
4269 ''', video_description)
4270 if mobj:
4271 release_year = mobj.group('release_year')
4272 release_date = mobj.group('release_date')
4273 if release_date:
4274 release_date = release_date.replace('-', '')
4275 if not release_year:
4276 release_year = release_date[:4]
4277 info.update({
4278 'album': mobj.group('album'.strip()),
4279 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4280 'track': mobj.group('track').strip(),
4281 'release_date': release_date,
4282 'release_year': int_or_none(release_year),
4283 })
4284
4285 initial_data = None
4286 if webpage:
4287 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4288 if not traverse_obj(initial_data, 'contents'):
4289 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4290 initial_data = None
4291 if not initial_data:
4292 query = {'videoId': video_id}
4293 query.update(self._get_checkok_params())
4294 initial_data = self._extract_response(
4295 item_id=video_id, ep='next', fatal=False,
4296 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4297 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4298 note='Downloading initial data API JSON')
4299
4300 info['comment_count'] = traverse_obj(initial_data, (
4301 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4302 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4303 ), (
4304 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4305 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4306 ), expected_type=self._get_count, get_all=False)
4307
4308 try: # This will error if there is no livechat
4309 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4310 except (KeyError, IndexError, TypeError):
4311 pass
4312 else:
4313 info.setdefault('subtitles', {})['live_chat'] = [{
4314 # url is needed to set cookies
4315 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4316 'video_id': video_id,
4317 'ext': 'json',
4318 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4319 else 'youtube_live_chat_replay'),
4320 }]
4321
4322 if initial_data:
4323 info['chapters'] = (
4324 self._extract_chapters_from_json(initial_data, duration)
4325 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4326 or self._extract_chapters_from_description(video_description, duration)
4327 or None)
4328
4329 info['heatmap'] = self._extract_heatmap_from_player_overlay(initial_data)
4330
4331 contents = traverse_obj(
4332 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4333 expected_type=list, default=[])
4334
4335 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4336 if vpir:
4337 stl = vpir.get('superTitleLink')
4338 if stl:
4339 stl = self._get_text(stl)
4340 if try_get(
4341 vpir,
4342 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4343 info['location'] = stl
4344 else:
4345 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4346 if mobj:
4347 info.update({
4348 'series': mobj.group(1),
4349 'season_number': int(mobj.group(2)),
4350 'episode_number': int(mobj.group(3)),
4351 })
4352 for tlb in (try_get(
4353 vpir,
4354 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4355 list) or []):
4356 tbrs = variadic(
4357 traverse_obj(
4358 tlb, ('toggleButtonRenderer', ...),
4359 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4360 for tbr in tbrs:
4361 for getter, regex in [(
4362 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4363 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4364 lambda x: x['accessibility'],
4365 lambda x: x['accessibilityData']['accessibilityData'],
4366 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4367 label = (try_get(tbr, getter, dict) or {}).get('label')
4368 if label:
4369 mobj = re.match(regex, label)
4370 if mobj:
4371 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4372 break
4373 sbr_tooltip = try_get(
4374 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4375 if sbr_tooltip:
4376 like_count, dislike_count = sbr_tooltip.split(' / ')
4377 info.update({
4378 'like_count': str_to_int(like_count),
4379 'dislike_count': str_to_int(dislike_count),
4380 })
4381 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4382 if vcr:
4383 vc = self._get_count(vcr, 'viewCount')
4384 # Upcoming premieres with waiting count are treated as live here
4385 if vcr.get('isLive'):
4386 info['concurrent_view_count'] = vc
4387 elif info.get('view_count') is None:
4388 info['view_count'] = vc
4389
4390 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4391 if vsir:
4392 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4393 info.update({
4394 'channel': self._get_text(vor, 'title'),
4395 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4396
4397 if not channel_handle:
4398 channel_handle = self.handle_from_url(
4399 traverse_obj(vor, (
4400 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4401 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4402 {str}), get_all=False))
4403
4404 rows = try_get(
4405 vsir,
4406 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4407 list) or []
4408 multiple_songs = False
4409 for row in rows:
4410 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4411 multiple_songs = True
4412 break
4413 for row in rows:
4414 mrr = row.get('metadataRowRenderer') or {}
4415 mrr_title = mrr.get('title')
4416 if not mrr_title:
4417 continue
4418 mrr_title = self._get_text(mrr, 'title')
4419 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4420 if mrr_title == 'License':
4421 info['license'] = mrr_contents_text
4422 elif not multiple_songs:
4423 if mrr_title == 'Album':
4424 info['album'] = mrr_contents_text
4425 elif mrr_title == 'Artist':
4426 info['artist'] = mrr_contents_text
4427 elif mrr_title == 'Song':
4428 info['track'] = mrr_contents_text
4429
4430 info.update({
4431 'uploader': info.get('channel'),
4432 'uploader_id': channel_handle,
4433 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4434 })
4435 # The upload date for scheduled, live and past live streams / premieres in microformats
4436 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4437 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4438 upload_date = (
4439 unified_strdate(get_first(microformats, 'uploadDate'))
4440 or unified_strdate(search_meta('uploadDate')))
4441 if not upload_date or (
4442 live_status in ('not_live', None)
4443 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4444 ):
4445 upload_date = strftime_or_none(
4446 self._parse_time_text(self._get_text(vpir, 'dateText')), '%Y%m%d') or upload_date
4447 info['upload_date'] = upload_date
4448
4449 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4450 v = info.get(s_k)
4451 if v:
4452 info[d_k] = v
4453
4454 badges = self._extract_badges(traverse_obj(contents, (..., 'videoPrimaryInfoRenderer'), get_all=False))
4455
4456 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4457 or get_first(video_details, 'isPrivate', expected_type=bool))
4458
4459 info['availability'] = (
4460 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4461 else self._availability(
4462 is_private=is_private,
4463 needs_premium=(
4464 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4465 or False if initial_data and is_private is not None else None),
4466 needs_subscription=(
4467 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4468 or False if initial_data and is_private is not None else None),
4469 needs_auth=info['age_limit'] >= 18,
4470 is_unlisted=None if is_private is None else (
4471 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4472 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4473
4474 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4475
4476 self.mark_watched(video_id, player_responses)
4477
4478 return info
4479
4480
4481 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4482 @staticmethod
4483 def passthrough_smuggled_data(func):
4484 def _smuggle(info, smuggled_data):
4485 if info.get('_type') not in ('url', 'url_transparent'):
4486 return info
4487 if smuggled_data.get('is_music_url'):
4488 parsed_url = urllib.parse.urlparse(info['url'])
4489 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4490 smuggled_data.pop('is_music_url')
4491 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4492 if smuggled_data:
4493 info['url'] = smuggle_url(info['url'], smuggled_data)
4494 return info
4495
4496 @functools.wraps(func)
4497 def wrapper(self, url):
4498 url, smuggled_data = unsmuggle_url(url, {})
4499 if self.is_music_url(url):
4500 smuggled_data['is_music_url'] = True
4501 info_dict = func(self, url, smuggled_data)
4502 if smuggled_data:
4503 _smuggle(info_dict, smuggled_data)
4504 if info_dict.get('entries'):
4505 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4506 return info_dict
4507 return wrapper
4508
4509 @staticmethod
4510 def _extract_basic_item_renderer(item):
4511 # Modified from _extract_grid_item_renderer
4512 known_basic_renderers = (
4513 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4514 )
4515 for key, renderer in item.items():
4516 if not isinstance(renderer, dict):
4517 continue
4518 elif key in known_basic_renderers:
4519 return renderer
4520 elif key.startswith('grid') and key.endswith('Renderer'):
4521 return renderer
4522
4523 def _extract_channel_renderer(self, renderer):
4524 channel_id = self.ucid_or_none(renderer['channelId'])
4525 title = self._get_text(renderer, 'title')
4526 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4527 # As of 2023-03-01 YouTube doesn't use the channel handles on these renderers yet.
4528 # However we can expect them to change that in the future.
4529 channel_handle = self.handle_from_url(
4530 traverse_obj(renderer, (
4531 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4532 ('browseEndpoint', 'canonicalBaseUrl')),
4533 {str}), get_all=False))
4534 return {
4535 '_type': 'url',
4536 'url': channel_url,
4537 'id': channel_id,
4538 'ie_key': YoutubeTabIE.ie_key(),
4539 'channel': title,
4540 'uploader': title,
4541 'channel_id': channel_id,
4542 'channel_url': channel_url,
4543 'title': title,
4544 'uploader_id': channel_handle,
4545 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4546 'channel_follower_count': self._get_count(renderer, 'subscriberCountText'),
4547 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4548 'playlist_count': self._get_count(renderer, 'videoCountText'),
4549 'description': self._get_text(renderer, 'descriptionSnippet'),
4550 }
4551
4552 def _grid_entries(self, grid_renderer):
4553 for item in grid_renderer['items']:
4554 if not isinstance(item, dict):
4555 continue
4556 renderer = self._extract_basic_item_renderer(item)
4557 if not isinstance(renderer, dict):
4558 continue
4559 title = self._get_text(renderer, 'title')
4560
4561 # playlist
4562 playlist_id = renderer.get('playlistId')
4563 if playlist_id:
4564 yield self.url_result(
4565 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4566 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4567 video_title=title)
4568 continue
4569 # video
4570 video_id = renderer.get('videoId')
4571 if video_id:
4572 yield self._extract_video(renderer)
4573 continue
4574 # channel
4575 channel_id = renderer.get('channelId')
4576 if channel_id:
4577 yield self._extract_channel_renderer(renderer)
4578 continue
4579 # generic endpoint URL support
4580 ep_url = urljoin('https://www.youtube.com/', try_get(
4581 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4582 str))
4583 if ep_url:
4584 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4585 if ie.suitable(ep_url):
4586 yield self.url_result(
4587 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4588 break
4589
4590 def _music_reponsive_list_entry(self, renderer):
4591 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4592 if video_id:
4593 title = traverse_obj(renderer, (
4594 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4595 'text', 'runs', 0, 'text'))
4596 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4597 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4598 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4599 if playlist_id:
4600 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4601 if video_id:
4602 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4603 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4604 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4605 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4606 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4607 if browse_id:
4608 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4609 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4610
4611 def _shelf_entries_from_content(self, shelf_renderer):
4612 content = shelf_renderer.get('content')
4613 if not isinstance(content, dict):
4614 return
4615 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4616 if renderer:
4617 # TODO: add support for nested playlists so each shelf is processed
4618 # as separate playlist
4619 # TODO: this includes only first N items
4620 yield from self._grid_entries(renderer)
4621 renderer = content.get('horizontalListRenderer')
4622 if renderer:
4623 # TODO
4624 pass
4625
4626 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4627 ep = try_get(
4628 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4629 str)
4630 shelf_url = urljoin('https://www.youtube.com', ep)
4631 if shelf_url:
4632 # Skipping links to another channels, note that checking for
4633 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4634 # will not work
4635 if skip_channels and '/channels?' in shelf_url:
4636 return
4637 title = self._get_text(shelf_renderer, 'title')
4638 yield self.url_result(shelf_url, video_title=title)
4639 # Shelf may not contain shelf URL, fallback to extraction from content
4640 yield from self._shelf_entries_from_content(shelf_renderer)
4641
4642 def _playlist_entries(self, video_list_renderer):
4643 for content in video_list_renderer['contents']:
4644 if not isinstance(content, dict):
4645 continue
4646 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4647 if not isinstance(renderer, dict):
4648 continue
4649 video_id = renderer.get('videoId')
4650 if not video_id:
4651 continue
4652 yield self._extract_video(renderer)
4653
4654 def _rich_entries(self, rich_grid_renderer):
4655 renderer = traverse_obj(
4656 rich_grid_renderer,
4657 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4658 video_id = renderer.get('videoId')
4659 if video_id:
4660 yield self._extract_video(renderer)
4661 return
4662 playlist_id = renderer.get('playlistId')
4663 if playlist_id:
4664 yield self.url_result(
4665 f'https://www.youtube.com/playlist?list={playlist_id}',
4666 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4667 video_title=self._get_text(renderer, 'title'))
4668 return
4669
4670 def _video_entry(self, video_renderer):
4671 video_id = video_renderer.get('videoId')
4672 if video_id:
4673 return self._extract_video(video_renderer)
4674
4675 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4676 url = urljoin('https://youtube.com', traverse_obj(
4677 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4678 if url:
4679 return self.url_result(
4680 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4681
4682 def _post_thread_entries(self, post_thread_renderer):
4683 post_renderer = try_get(
4684 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4685 if not post_renderer:
4686 return
4687 # video attachment
4688 video_renderer = try_get(
4689 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4690 video_id = video_renderer.get('videoId')
4691 if video_id:
4692 entry = self._extract_video(video_renderer)
4693 if entry:
4694 yield entry
4695 # playlist attachment
4696 playlist_id = try_get(
4697 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4698 if playlist_id:
4699 yield self.url_result(
4700 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4701 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4702 # inline video links
4703 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4704 for run in runs:
4705 if not isinstance(run, dict):
4706 continue
4707 ep_url = try_get(
4708 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4709 if not ep_url:
4710 continue
4711 if not YoutubeIE.suitable(ep_url):
4712 continue
4713 ep_video_id = YoutubeIE._match_id(ep_url)
4714 if video_id == ep_video_id:
4715 continue
4716 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4717
4718 def _post_thread_continuation_entries(self, post_thread_continuation):
4719 contents = post_thread_continuation.get('contents')
4720 if not isinstance(contents, list):
4721 return
4722 for content in contents:
4723 renderer = content.get('backstagePostThreadRenderer')
4724 if isinstance(renderer, dict):
4725 yield from self._post_thread_entries(renderer)
4726 continue
4727 renderer = content.get('videoRenderer')
4728 if isinstance(renderer, dict):
4729 yield self._video_entry(renderer)
4730
4731 r''' # unused
4732 def _rich_grid_entries(self, contents):
4733 for content in contents:
4734 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4735 if video_renderer:
4736 entry = self._video_entry(video_renderer)
4737 if entry:
4738 yield entry
4739 '''
4740
4741 def _report_history_entries(self, renderer):
4742 for url in traverse_obj(renderer, (
4743 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4744 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4745 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4746 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4747
4748 def _extract_entries(self, parent_renderer, continuation_list):
4749 # continuation_list is modified in-place with continuation_list = [continuation_token]
4750 continuation_list[:] = [None]
4751 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4752 for content in contents:
4753 if not isinstance(content, dict):
4754 continue
4755 is_renderer = traverse_obj(
4756 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4757 expected_type=dict)
4758 if not is_renderer:
4759 if content.get('richItemRenderer'):
4760 for entry in self._rich_entries(content['richItemRenderer']):
4761 yield entry
4762 continuation_list[0] = self._extract_continuation(parent_renderer)
4763 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4764 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4765 yield from self._report_history_entries(table)
4766 continuation_list[0] = self._extract_continuation(table)
4767 continue
4768
4769 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4770 for isr_content in isr_contents:
4771 if not isinstance(isr_content, dict):
4772 continue
4773
4774 known_renderers = {
4775 'playlistVideoListRenderer': self._playlist_entries,
4776 'gridRenderer': self._grid_entries,
4777 'reelShelfRenderer': self._grid_entries,
4778 'shelfRenderer': self._shelf_entries,
4779 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4780 'backstagePostThreadRenderer': self._post_thread_entries,
4781 'videoRenderer': lambda x: [self._video_entry(x)],
4782 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4783 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4784 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4785 }
4786 for key, renderer in isr_content.items():
4787 if key not in known_renderers:
4788 continue
4789 for entry in known_renderers[key](renderer):
4790 if entry:
4791 yield entry
4792 continuation_list[0] = self._extract_continuation(renderer)
4793 break
4794
4795 if not continuation_list[0]:
4796 continuation_list[0] = self._extract_continuation(is_renderer)
4797
4798 if not continuation_list[0]:
4799 continuation_list[0] = self._extract_continuation(parent_renderer)
4800
4801 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4802 continuation_list = [None]
4803 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4804 tab_content = try_get(tab, lambda x: x['content'], dict)
4805 if not tab_content:
4806 return
4807 parent_renderer = (
4808 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4809 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4810 yield from extract_entries(parent_renderer)
4811 continuation = continuation_list[0]
4812
4813 for page_num in itertools.count(1):
4814 if not continuation:
4815 break
4816 headers = self.generate_api_headers(
4817 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4818 response = self._extract_response(
4819 item_id=f'{item_id} page {page_num}',
4820 query=continuation, headers=headers, ytcfg=ytcfg,
4821 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4822
4823 if not response:
4824 break
4825 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4826 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4827 visitor_data = self._extract_visitor_data(response) or visitor_data
4828
4829 known_renderers = {
4830 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4831 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4832 'gridVideoRenderer': (self._grid_entries, 'items'),
4833 'gridChannelRenderer': (self._grid_entries, 'items'),
4834 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4835 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4836 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4837 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4838 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4839 'playlistVideoListContinuation': (self._playlist_entries, None),
4840 'gridContinuation': (self._grid_entries, None),
4841 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4842 'sectionListContinuation': (extract_entries, None), # for feeds
4843 }
4844
4845 continuation_items = traverse_obj(response, (
4846 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4847 'appendContinuationItemsAction', 'continuationItems'
4848 ), 'continuationContents', get_all=False)
4849 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4850
4851 video_items_renderer = None
4852 for key in continuation_item.keys():
4853 if key not in known_renderers:
4854 continue
4855 func, parent_key = known_renderers[key]
4856 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4857 continuation_list = [None]
4858 yield from func(video_items_renderer)
4859 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4860
4861 if not video_items_renderer:
4862 break
4863
4864 @staticmethod
4865 def _extract_selected_tab(tabs, fatal=True):
4866 for tab_renderer in tabs:
4867 if tab_renderer.get('selected'):
4868 return tab_renderer
4869 if fatal:
4870 raise ExtractorError('Unable to find selected tab')
4871
4872 @staticmethod
4873 def _extract_tab_renderers(response):
4874 return traverse_obj(
4875 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4876
4877 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4878 metadata = self._extract_metadata_from_tabs(item_id, data)
4879
4880 selected_tab = self._extract_selected_tab(tabs)
4881 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4882 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
4883
4884 return self.playlist_result(
4885 self._entries(
4886 selected_tab, metadata['id'], ytcfg,
4887 self._extract_account_syncid(ytcfg, data),
4888 self._extract_visitor_data(data, ytcfg)),
4889 **metadata)
4890
4891 def _extract_metadata_from_tabs(self, item_id, data):
4892 info = {'id': item_id}
4893
4894 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
4895 if metadata_renderer:
4896 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
4897 ('channelUrl', {self.ucid_from_url}))
4898 info.update({
4899 'channel': metadata_renderer.get('title'),
4900 'channel_id': channel_id,
4901 })
4902 if info['channel_id']:
4903 info['id'] = info['channel_id']
4904 else:
4905 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
4906
4907 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4908 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4909 def _get_uncropped(url):
4910 return url_or_none((url or '').split('=')[0] + '=s0')
4911
4912 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
4913 if avatar_thumbnails:
4914 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4915 if uncropped_avatar:
4916 avatar_thumbnails.append({
4917 'url': uncropped_avatar,
4918 'id': 'avatar_uncropped',
4919 'preference': 1
4920 })
4921
4922 channel_banners = self._extract_thumbnails(
4923 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
4924 for banner in channel_banners:
4925 banner['preference'] = -10
4926
4927 if channel_banners:
4928 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4929 if uncropped_banner:
4930 channel_banners.append({
4931 'url': uncropped_banner,
4932 'id': 'banner_uncropped',
4933 'preference': -5
4934 })
4935
4936 # Deprecated - remove primary_sidebar_renderer when layout discontinued
4937 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4938 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
4939
4940 primary_thumbnails = self._extract_thumbnails(
4941 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4942 playlist_thumbnails = self._extract_thumbnails(
4943 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
4944
4945 info.update({
4946 'title': (traverse_obj(metadata_renderer, 'title')
4947 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
4948 or info['id']),
4949 'availability': self._extract_availability(data),
4950 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4951 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
4952 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
4953 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
4954 })
4955
4956 channel_handle = (
4957 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
4958 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
4959
4960 if channel_handle:
4961 info.update({
4962 'uploader_id': channel_handle,
4963 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4964 })
4965 # Playlist stats is a text runs array containing [video count, view count, last updated].
4966 # last updated or (view count and last updated) may be missing.
4967 playlist_stats = get_first(
4968 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
4969
4970 last_updated_unix = self._parse_time_text(
4971 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
4972 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
4973 info['modified_date'] = strftime_or_none(last_updated_unix, '%Y%m%d')
4974
4975 info['view_count'] = self._get_count(playlist_stats, 1)
4976 if info['view_count'] is None: # 0 is allowed
4977 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
4978 if info['view_count'] is None:
4979 info['view_count'] = self._get_count(data, (
4980 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
4981 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
4982
4983 info['playlist_count'] = self._get_count(playlist_stats, 0)
4984 if info['playlist_count'] is None: # 0 is allowed
4985 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
4986
4987 if not info.get('channel_id'):
4988 owner = traverse_obj(playlist_header_renderer, 'ownerText')
4989 if not owner: # Deprecated
4990 owner = traverse_obj(
4991 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
4992 ('videoOwner', 'videoOwnerRenderer', 'title'))
4993 owner_text = self._get_text(owner)
4994 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
4995 info.update({
4996 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
4997 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
4998 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
4999 })
5000
5001 info.update({
5002 'uploader': info['channel'],
5003 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5004 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5005 })
5006
5007 return info
5008
5009 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5010 first_id = last_id = response = None
5011 for page_num in itertools.count(1):
5012 videos = list(self._playlist_entries(playlist))
5013 if not videos:
5014 return
5015 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5016 if start >= len(videos):
5017 return
5018 yield from videos[start:]
5019 first_id = first_id or videos[0]['id']
5020 last_id = videos[-1]['id']
5021 watch_endpoint = try_get(
5022 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5023 headers = self.generate_api_headers(
5024 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5025 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5026 query = {
5027 'playlistId': playlist_id,
5028 'videoId': watch_endpoint.get('videoId') or last_id,
5029 'index': watch_endpoint.get('index') or len(videos),
5030 'params': watch_endpoint.get('params') or 'OAE%3D'
5031 }
5032 response = self._extract_response(
5033 item_id='%s page %d' % (playlist_id, page_num),
5034 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5035 check_get_keys='contents'
5036 )
5037 playlist = try_get(
5038 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5039
5040 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5041 title = playlist.get('title') or try_get(
5042 data, lambda x: x['titleText']['simpleText'], str)
5043 playlist_id = playlist.get('playlistId') or item_id
5044
5045 # Delegating everything except mix playlists to regular tab-based playlist URL
5046 playlist_url = urljoin(url, try_get(
5047 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5048 str))
5049
5050 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5051 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5052 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5053
5054 if playlist_url and playlist_url != url and not is_known_unviewable:
5055 return self.url_result(
5056 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5057 video_title=title)
5058
5059 return self.playlist_result(
5060 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5061 playlist_id=playlist_id, playlist_title=title)
5062
5063 def _extract_availability(self, data):
5064 """
5065 Gets the availability of a given playlist/tab.
5066 Note: Unless YouTube tells us explicitly, we do not assume it is public
5067 @param data: response
5068 """
5069 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5070 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5071 player_header_privacy = playlist_header_renderer.get('privacy')
5072
5073 badges = self._extract_badges(sidebar_renderer)
5074
5075 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5076 privacy_setting_icon = get_first(
5077 (playlist_header_renderer, sidebar_renderer),
5078 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5079 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5080 expected_type=str)
5081
5082 microformats_is_unlisted = traverse_obj(
5083 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5084
5085 return (
5086 'public' if (
5087 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5088 or player_header_privacy == 'PUBLIC'
5089 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5090 else self._availability(
5091 is_private=(
5092 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5093 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5094 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5095 is_unlisted=(
5096 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5097 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5098 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5099 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5100 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5101 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5102 needs_auth=False))
5103
5104 @staticmethod
5105 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5106 sidebar_renderer = try_get(
5107 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5108 for item in sidebar_renderer:
5109 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5110 if renderer:
5111 return renderer
5112
5113 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5114 """
5115 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5116 """
5117 is_playlist = bool(traverse_obj(
5118 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5119 if not is_playlist:
5120 return
5121 headers = self.generate_api_headers(
5122 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5123 visitor_data=self._extract_visitor_data(data, ytcfg))
5124 query = {
5125 'params': 'wgYCCAA=',
5126 'browseId': f'VL{item_id}'
5127 }
5128 return self._extract_response(
5129 item_id=item_id, headers=headers, query=query,
5130 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5131 note='Redownloading playlist API JSON with unavailable videos')
5132
5133 @functools.cached_property
5134 def skip_webpage(self):
5135 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5136
5137 def _extract_webpage(self, url, item_id, fatal=True):
5138 webpage, data = None, None
5139 for retry in self.RetryManager(fatal=fatal):
5140 try:
5141 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5142 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5143 except ExtractorError as e:
5144 if isinstance(e.cause, network_exceptions):
5145 if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
5146 retry.error = e
5147 continue
5148 self._error_or_warning(e, fatal=fatal)
5149 break
5150
5151 try:
5152 self._extract_and_report_alerts(data)
5153 except ExtractorError as e:
5154 self._error_or_warning(e, fatal=fatal)
5155 break
5156
5157 # Sometimes youtube returns a webpage with incomplete ytInitialData
5158 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5159 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5160 retry.error = ExtractorError('Incomplete yt initial data received')
5161 continue
5162
5163 return webpage, data
5164
5165 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5166 """Use if failed to extract ytcfg (and data) from initial webpage"""
5167 if not ytcfg and self.is_authenticated:
5168 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5169 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5170 raise ExtractorError(
5171 f'{msg}. If you are not downloading private content, or '
5172 'your cookies are only for the first account and channel,'
5173 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5174 expected=True)
5175 self.report_warning(msg, only_once=True)
5176
5177 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5178 data = None
5179 if not self.skip_webpage:
5180 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5181 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5182 # Reject webpage data if redirected to home page without explicitly requesting
5183 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5184 if (url != 'https://www.youtube.com/feed/recommended'
5185 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5186 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5187 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5188 if fatal:
5189 raise ExtractorError(msg, expected=True)
5190 self.report_warning(msg, only_once=True)
5191 if not data:
5192 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5193 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5194 return data, ytcfg
5195
5196 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5197 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5198 resolve_response = self._extract_response(
5199 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5200 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5201 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5202 for ep_key, ep in endpoints.items():
5203 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5204 if params:
5205 return self._extract_response(
5206 item_id=item_id, query=params, ep=ep, headers=headers,
5207 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5208 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5209 err_note = 'Failed to resolve url (does the playlist exist?)'
5210 if fatal:
5211 raise ExtractorError(err_note, expected=True)
5212 self.report_warning(err_note, item_id)
5213
5214 _SEARCH_PARAMS = None
5215
5216 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5217 data = {'query': query}
5218 if params is NO_DEFAULT:
5219 params = self._SEARCH_PARAMS
5220 if params:
5221 data['params'] = params
5222
5223 content_keys = (
5224 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5225 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5226 # ytmusic search
5227 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5228 ('continuationContents', ),
5229 )
5230 display_id = f'query "{query}"'
5231 check_get_keys = tuple({keys[0] for keys in content_keys})
5232 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5233 self._report_playlist_authcheck(ytcfg, fatal=False)
5234
5235 continuation_list = [None]
5236 search = None
5237 for page_num in itertools.count(1):
5238 data.update(continuation_list[0] or {})
5239 headers = self.generate_api_headers(
5240 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5241 search = self._extract_response(
5242 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5243 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5244 slr_contents = traverse_obj(search, *content_keys)
5245 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5246 if not continuation_list[0]:
5247 break
5248
5249
5250 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5251 IE_DESC = 'YouTube Tabs'
5252 _VALID_URL = r'''(?x:
5253 https?://
5254 (?!consent\.)(?:\w+\.)?
5255 (?:
5256 youtube(?:kids)?\.com|
5257 %(invidious)s
5258 )/
5259 (?:
5260 (?P<channel_type>channel|c|user|browse)/|
5261 (?P<not_channel>
5262 feed/|hashtag/|
5263 (?:playlist|watch)\?.*?\blist=
5264 )|
5265 (?!(?:%(reserved_names)s)\b) # Direct URLs
5266 )
5267 (?P<id>[^/?\#&]+)
5268 )''' % {
5269 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5270 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5271 }
5272 IE_NAME = 'youtube:tab'
5273
5274 _TESTS = [{
5275 'note': 'playlists, multipage',
5276 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5277 'playlist_mincount': 94,
5278 'info_dict': {
5279 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5280 'title': 'Igor Kleiner - Playlists',
5281 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5282 'uploader': 'Igor Kleiner',
5283 'uploader_id': '@IgorDataScience',
5284 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5285 'channel': 'Igor Kleiner',
5286 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5287 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5288 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5289 'channel_follower_count': int
5290 },
5291 }, {
5292 'note': 'playlists, multipage, different order',
5293 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5294 'playlist_mincount': 94,
5295 'info_dict': {
5296 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5297 'title': 'Igor Kleiner - Playlists',
5298 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5299 'uploader': 'Igor Kleiner',
5300 'uploader_id': '@IgorDataScience',
5301 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5302 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5303 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5304 'channel': 'Igor Kleiner',
5305 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5306 'channel_follower_count': int
5307 },
5308 }, {
5309 'note': 'playlists, series',
5310 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5311 'playlist_mincount': 5,
5312 'info_dict': {
5313 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5314 'title': '3Blue1Brown - Playlists',
5315 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5316 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5317 'channel': '3Blue1Brown',
5318 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5319 'uploader_id': '@3blue1brown',
5320 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5321 'uploader': '3Blue1Brown',
5322 'tags': ['Mathematics'],
5323 'channel_follower_count': int
5324 },
5325 }, {
5326 'note': 'playlists, singlepage',
5327 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5328 'playlist_mincount': 4,
5329 'info_dict': {
5330 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5331 'title': 'ThirstForScience - Playlists',
5332 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5333 'uploader': 'ThirstForScience',
5334 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5335 'uploader_id': '@ThirstForScience',
5336 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5337 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5338 'tags': 'count:13',
5339 'channel': 'ThirstForScience',
5340 'channel_follower_count': int
5341 }
5342 }, {
5343 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5344 'only_matching': True,
5345 }, {
5346 'note': 'basic, single video playlist',
5347 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5348 'info_dict': {
5349 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5350 'title': 'youtube-dl public playlist',
5351 'description': '',
5352 'tags': [],
5353 'view_count': int,
5354 'modified_date': '20201130',
5355 'channel': 'Sergey M.',
5356 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5357 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5358 'availability': 'public',
5359 'uploader': 'Sergey M.',
5360 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5361 'uploader_id': '@sergeym.6173',
5362 },
5363 'playlist_count': 1,
5364 }, {
5365 'note': 'empty playlist',
5366 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5367 'info_dict': {
5368 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5369 'title': 'youtube-dl empty playlist',
5370 'tags': [],
5371 'channel': 'Sergey M.',
5372 'description': '',
5373 'modified_date': '20160902',
5374 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5375 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5376 'availability': 'public',
5377 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5378 'uploader_id': '@sergeym.6173',
5379 'uploader': 'Sergey M.',
5380 },
5381 'playlist_count': 0,
5382 }, {
5383 'note': 'Home tab',
5384 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5385 'info_dict': {
5386 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5387 'title': 'lex will - Home',
5388 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5389 'uploader': 'lex will',
5390 'uploader_id': '@lexwill718',
5391 'channel': 'lex will',
5392 'tags': ['bible', 'history', 'prophesy'],
5393 'uploader_url': 'https://www.youtube.com/@lexwill718',
5394 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5395 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5396 'channel_follower_count': int
5397 },
5398 'playlist_mincount': 2,
5399 }, {
5400 'note': 'Videos tab',
5401 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5402 'info_dict': {
5403 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5404 'title': 'lex will - Videos',
5405 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5406 'uploader': 'lex will',
5407 'uploader_id': '@lexwill718',
5408 'tags': ['bible', 'history', 'prophesy'],
5409 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5410 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5411 'uploader_url': 'https://www.youtube.com/@lexwill718',
5412 'channel': 'lex will',
5413 'channel_follower_count': int
5414 },
5415 'playlist_mincount': 975,
5416 }, {
5417 'note': 'Videos tab, sorted by popular',
5418 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5419 'info_dict': {
5420 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5421 'title': 'lex will - Videos',
5422 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5423 'uploader': 'lex will',
5424 'uploader_id': '@lexwill718',
5425 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5426 'uploader_url': 'https://www.youtube.com/@lexwill718',
5427 'channel': 'lex will',
5428 'tags': ['bible', 'history', 'prophesy'],
5429 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5430 'channel_follower_count': int
5431 },
5432 'playlist_mincount': 199,
5433 }, {
5434 'note': 'Playlists tab',
5435 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5436 'info_dict': {
5437 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5438 'title': 'lex will - Playlists',
5439 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5440 'uploader': 'lex will',
5441 'uploader_id': '@lexwill718',
5442 'uploader_url': 'https://www.youtube.com/@lexwill718',
5443 'channel': 'lex will',
5444 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5445 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5446 'tags': ['bible', 'history', 'prophesy'],
5447 'channel_follower_count': int
5448 },
5449 'playlist_mincount': 17,
5450 }, {
5451 'note': 'Community tab',
5452 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5453 'info_dict': {
5454 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5455 'title': 'lex will - Community',
5456 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5457 'channel': 'lex will',
5458 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5459 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5460 'tags': ['bible', 'history', 'prophesy'],
5461 'channel_follower_count': int,
5462 'uploader_url': 'https://www.youtube.com/@lexwill718',
5463 'uploader_id': '@lexwill718',
5464 'uploader': 'lex will',
5465 },
5466 'playlist_mincount': 18,
5467 }, {
5468 'note': 'Channels tab',
5469 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5470 'info_dict': {
5471 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5472 'title': 'lex will - Channels',
5473 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5474 'channel': 'lex will',
5475 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5476 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5477 'tags': ['bible', 'history', 'prophesy'],
5478 'channel_follower_count': int,
5479 'uploader_url': 'https://www.youtube.com/@lexwill718',
5480 'uploader_id': '@lexwill718',
5481 'uploader': 'lex will',
5482 },
5483 'playlist_mincount': 12,
5484 }, {
5485 'note': 'Search tab',
5486 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5487 'playlist_mincount': 40,
5488 'info_dict': {
5489 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5490 'title': '3Blue1Brown - Search - linear algebra',
5491 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5492 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5493 'tags': ['Mathematics'],
5494 'channel': '3Blue1Brown',
5495 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5496 'channel_follower_count': int,
5497 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5498 'uploader_id': '@3blue1brown',
5499 'uploader': '3Blue1Brown',
5500 },
5501 }, {
5502 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5503 'only_matching': True,
5504 }, {
5505 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5506 'only_matching': True,
5507 }, {
5508 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5509 'only_matching': True,
5510 }, {
5511 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5512 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5513 'info_dict': {
5514 'title': '29C3: Not my department',
5515 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5516 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5517 'tags': [],
5518 'view_count': int,
5519 'modified_date': '20150605',
5520 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5521 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5522 'channel': 'Christiaan008',
5523 'availability': 'public',
5524 'uploader_id': '@ChRiStIaAn008',
5525 'uploader': 'Christiaan008',
5526 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5527 },
5528 'playlist_count': 96,
5529 }, {
5530 'note': 'Large playlist',
5531 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5532 'info_dict': {
5533 'title': 'Uploads from Cauchemar',
5534 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5535 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5536 'tags': [],
5537 'modified_date': r're:\d{8}',
5538 'channel': 'Cauchemar',
5539 'view_count': int,
5540 'description': '',
5541 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5542 'availability': 'public',
5543 'uploader_id': '@Cauchemar89',
5544 'uploader': 'Cauchemar',
5545 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5546 },
5547 'playlist_mincount': 1123,
5548 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5549 }, {
5550 'note': 'even larger playlist, 8832 videos',
5551 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5552 'only_matching': True,
5553 }, {
5554 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5555 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5556 'info_dict': {
5557 'title': 'Uploads from Interstellar Movie',
5558 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5559 'tags': [],
5560 'view_count': int,
5561 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5562 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5563 'channel': 'Interstellar Movie',
5564 'description': '',
5565 'modified_date': r're:\d{8}',
5566 'availability': 'public',
5567 'uploader_id': '@InterstellarMovie',
5568 'uploader': 'Interstellar Movie',
5569 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5570 },
5571 'playlist_mincount': 21,
5572 }, {
5573 'note': 'Playlist with "show unavailable videos" button',
5574 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5575 'info_dict': {
5576 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5577 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5578 'view_count': int,
5579 'channel': 'Phim Siêu Nhân Nhật Bản',
5580 'tags': [],
5581 'description': '',
5582 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5583 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5584 'modified_date': r're:\d{8}',
5585 'availability': 'public',
5586 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5587 'uploader_id': '@phimsieunhannhatban',
5588 'uploader': 'Phim Siêu Nhân Nhật Bản',
5589 },
5590 'playlist_mincount': 200,
5591 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5592 }, {
5593 'note': 'Playlist with unavailable videos in page 7',
5594 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5595 'info_dict': {
5596 'title': 'Uploads from BlankTV',
5597 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5598 'channel': 'BlankTV',
5599 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5600 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5601 'view_count': int,
5602 'tags': [],
5603 'modified_date': r're:\d{8}',
5604 'description': '',
5605 'availability': 'public',
5606 'uploader_id': '@blanktv',
5607 'uploader': 'BlankTV',
5608 'uploader_url': 'https://www.youtube.com/@blanktv',
5609 },
5610 'playlist_mincount': 1000,
5611 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5612 }, {
5613 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5614 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5615 'info_dict': {
5616 'title': 'Data Analysis with Dr Mike Pound',
5617 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5618 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5619 'tags': [],
5620 'view_count': int,
5621 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5622 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5623 'channel': 'Computerphile',
5624 'availability': 'public',
5625 'modified_date': '20190712',
5626 'uploader_id': '@Computerphile',
5627 'uploader': 'Computerphile',
5628 'uploader_url': 'https://www.youtube.com/@Computerphile',
5629 },
5630 'playlist_mincount': 11,
5631 }, {
5632 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5633 'only_matching': True,
5634 }, {
5635 'note': 'Playlist URL that does not actually serve a playlist',
5636 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5637 'info_dict': {
5638 'id': 'FqZTN594JQw',
5639 'ext': 'webm',
5640 'title': "Smiley's People 01 detective, Adventure Series, Action",
5641 'upload_date': '20150526',
5642 'license': 'Standard YouTube License',
5643 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5644 'categories': ['People & Blogs'],
5645 'tags': list,
5646 'view_count': int,
5647 'like_count': int,
5648 },
5649 'params': {
5650 'skip_download': True,
5651 },
5652 'skip': 'This video is not available.',
5653 'add_ie': [YoutubeIE.ie_key()],
5654 }, {
5655 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5656 'only_matching': True,
5657 }, {
5658 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5659 'only_matching': True,
5660 }, {
5661 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5662 'info_dict': {
5663 'id': 'AlTsmyW4auo', # This will keep changing
5664 'ext': 'mp4',
5665 'title': str,
5666 'upload_date': r're:\d{8}',
5667 'description': str,
5668 'categories': ['News & Politics'],
5669 'tags': list,
5670 'like_count': int,
5671 'release_timestamp': int,
5672 'channel': 'Sky News',
5673 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5674 'age_limit': 0,
5675 'view_count': int,
5676 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5677 'playable_in_embed': True,
5678 'release_date': r're:\d+',
5679 'availability': 'public',
5680 'live_status': 'is_live',
5681 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5682 'channel_follower_count': int,
5683 'concurrent_view_count': int,
5684 'uploader_url': 'https://www.youtube.com/@SkyNews',
5685 'uploader_id': '@SkyNews',
5686 'uploader': 'Sky News',
5687 },
5688 'params': {
5689 'skip_download': True,
5690 },
5691 'expected_warnings': ['Ignoring subtitle tracks found in '],
5692 }, {
5693 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5694 'info_dict': {
5695 'id': 'a48o2S1cPoo',
5696 'ext': 'mp4',
5697 'title': 'The Young Turks - Live Main Show',
5698 'upload_date': '20150715',
5699 'license': 'Standard YouTube License',
5700 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5701 'categories': ['News & Politics'],
5702 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5703 'like_count': int,
5704 },
5705 'params': {
5706 'skip_download': True,
5707 },
5708 'only_matching': True,
5709 }, {
5710 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5711 'only_matching': True,
5712 }, {
5713 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5714 'only_matching': True,
5715 }, {
5716 'note': 'A channel that is not live. Should raise error',
5717 'url': 'https://www.youtube.com/user/numberphile/live',
5718 'only_matching': True,
5719 }, {
5720 'url': 'https://www.youtube.com/feed/trending',
5721 'only_matching': True,
5722 }, {
5723 'url': 'https://www.youtube.com/feed/library',
5724 'only_matching': True,
5725 }, {
5726 'url': 'https://www.youtube.com/feed/history',
5727 'only_matching': True,
5728 }, {
5729 'url': 'https://www.youtube.com/feed/subscriptions',
5730 'only_matching': True,
5731 }, {
5732 'url': 'https://www.youtube.com/feed/watch_later',
5733 'only_matching': True,
5734 }, {
5735 'note': 'Recommended - redirects to home page.',
5736 'url': 'https://www.youtube.com/feed/recommended',
5737 'only_matching': True,
5738 }, {
5739 'note': 'inline playlist with not always working continuations',
5740 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5741 'only_matching': True,
5742 }, {
5743 'url': 'https://www.youtube.com/course',
5744 'only_matching': True,
5745 }, {
5746 'url': 'https://www.youtube.com/zsecurity',
5747 'only_matching': True,
5748 }, {
5749 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5750 'only_matching': True,
5751 }, {
5752 'url': 'https://www.youtube.com/TheYoungTurks/live',
5753 'only_matching': True,
5754 }, {
5755 'url': 'https://www.youtube.com/hashtag/cctv9',
5756 'info_dict': {
5757 'id': 'cctv9',
5758 'title': '#cctv9',
5759 'tags': [],
5760 },
5761 'playlist_mincount': 300, # not consistent but should be over 300
5762 }, {
5763 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5764 'only_matching': True,
5765 }, {
5766 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5767 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5768 'only_matching': True
5769 }, {
5770 'note': '/browse/ should redirect to /channel/',
5771 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5772 'only_matching': True
5773 }, {
5774 'note': 'VLPL, should redirect to playlist?list=PL...',
5775 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5776 'info_dict': {
5777 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5778 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5779 'title': 'NCS : All Releases 💿',
5780 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5781 'modified_date': r're:\d{8}',
5782 'view_count': int,
5783 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5784 'tags': [],
5785 'channel': 'NoCopyrightSounds',
5786 'availability': 'public',
5787 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5788 'uploader': 'NoCopyrightSounds',
5789 'uploader_id': '@NoCopyrightSounds',
5790 },
5791 'playlist_mincount': 166,
5792 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5793 }, {
5794 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5795 'note': 'Topic, should redirect to playlist?list=UU...',
5796 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5797 'info_dict': {
5798 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5799 'title': 'Uploads from Royalty Free Music - Topic',
5800 'tags': [],
5801 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5802 'channel': 'Royalty Free Music - Topic',
5803 'view_count': int,
5804 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5805 'modified_date': r're:\d{8}',
5806 'description': '',
5807 'availability': 'public',
5808 'uploader': 'Royalty Free Music - Topic',
5809 },
5810 'playlist_mincount': 101,
5811 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5812 }, {
5813 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5814 # Treat as a general feed
5815 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5816 'info_dict': {
5817 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5818 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5819 'tags': [],
5820 },
5821 'playlist_mincount': 9,
5822 }, {
5823 'note': 'Youtube music Album',
5824 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5825 'info_dict': {
5826 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5827 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5828 'tags': [],
5829 'view_count': int,
5830 'description': '',
5831 'availability': 'unlisted',
5832 'modified_date': r're:\d{8}',
5833 },
5834 'playlist_count': 50,
5835 'expected_warnings': ['YouTube Music is not directly supported'],
5836 }, {
5837 'note': 'unlisted single video playlist',
5838 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5839 'info_dict': {
5840 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5841 'title': 'yt-dlp unlisted playlist test',
5842 'availability': 'unlisted',
5843 'tags': [],
5844 'modified_date': '20220418',
5845 'channel': 'colethedj',
5846 'view_count': int,
5847 'description': '',
5848 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5849 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5850 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5851 'uploader_id': '@colethedj1894',
5852 'uploader': 'colethedj',
5853 },
5854 'playlist_count': 1,
5855 }, {
5856 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5857 'url': 'https://www.youtube.com/feed/recommended',
5858 'info_dict': {
5859 'id': 'recommended',
5860 'title': 'recommended',
5861 'tags': [],
5862 },
5863 'playlist_mincount': 50,
5864 'params': {
5865 'skip_download': True,
5866 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5867 },
5868 }, {
5869 'note': 'API Fallback: /videos tab, sorted by oldest first',
5870 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5871 'info_dict': {
5872 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5873 'title': 'Cody\'sLab - Videos',
5874 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5875 'channel': 'Cody\'sLab',
5876 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5877 'tags': [],
5878 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5879 'channel_follower_count': int
5880 },
5881 'playlist_mincount': 650,
5882 'params': {
5883 'skip_download': True,
5884 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5885 },
5886 'skip': 'Query for sorting no longer works',
5887 }, {
5888 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5889 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5890 'info_dict': {
5891 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5892 'title': 'Uploads from Royalty Free Music - Topic',
5893 'modified_date': r're:\d{8}',
5894 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5895 'description': '',
5896 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5897 'tags': [],
5898 'channel': 'Royalty Free Music - Topic',
5899 'view_count': int,
5900 'availability': 'public',
5901 'uploader': 'Royalty Free Music - Topic',
5902 },
5903 'playlist_mincount': 101,
5904 'params': {
5905 'skip_download': True,
5906 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5907 },
5908 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5909 }, {
5910 'note': 'non-standard redirect to regional channel',
5911 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5912 'only_matching': True
5913 }, {
5914 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5915 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5916 'info_dict': {
5917 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5918 'modified_date': '20220407',
5919 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5920 'tags': [],
5921 'availability': 'unlisted',
5922 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5923 'channel': 'pukkandan',
5924 'description': 'Test for collaborative playlist',
5925 'title': 'yt-dlp test - collaborative playlist',
5926 'view_count': int,
5927 'uploader_url': 'https://www.youtube.com/@pukkandan',
5928 'uploader_id': '@pukkandan',
5929 'uploader': 'pukkandan',
5930 },
5931 'playlist_mincount': 2
5932 }, {
5933 'note': 'translated tab name',
5934 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
5935 'info_dict': {
5936 'id': 'UCiu-3thuViMebBjw_5nWYrA',
5937 'tags': [],
5938 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5939 'description': 'test description',
5940 'title': 'cole-dlp-test-acc - 再生リスト',
5941 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5942 'channel': 'cole-dlp-test-acc',
5943 'uploader_url': 'https://www.youtube.com/@coletdjnz',
5944 'uploader_id': '@coletdjnz',
5945 'uploader': 'cole-dlp-test-acc',
5946 },
5947 'playlist_mincount': 1,
5948 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5949 'expected_warnings': ['Preferring "ja"'],
5950 }, {
5951 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
5952 'note': 'preferred lang set with playlist with translated video titles',
5953 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
5954 'info_dict': {
5955 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
5956 'tags': [],
5957 'view_count': int,
5958 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5959 'channel': 'cole-dlp-test-acc',
5960 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5961 'description': 'test',
5962 'title': 'dlp test playlist',
5963 'availability': 'public',
5964 'uploader_url': 'https://www.youtube.com/@coletdjnz',
5965 'uploader_id': '@coletdjnz',
5966 'uploader': 'cole-dlp-test-acc',
5967 },
5968 'playlist_mincount': 1,
5969 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5970 'expected_warnings': ['Preferring "ja"'],
5971 }, {
5972 # shorts audio pivot for 2GtVksBMYFM.
5973 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
5974 'info_dict': {
5975 'id': 'sfv_audio_pivot',
5976 'title': 'sfv_audio_pivot',
5977 'tags': [],
5978 },
5979 'playlist_mincount': 50,
5980
5981 }, {
5982 # Channel with a real live tab (not to be mistaken with streams tab)
5983 # Do not treat like it should redirect to live stream
5984 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
5985 'info_dict': {
5986 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
5987 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
5988 'tags': [],
5989 },
5990 'playlist_mincount': 20,
5991 }, {
5992 # Tab name is not the same as tab id
5993 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
5994 'info_dict': {
5995 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
5996 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
5997 'tags': [],
5998 },
5999 'playlist_mincount': 8,
6000 }, {
6001 # Home tab id is literally home. Not to get mistaken with featured
6002 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6003 'info_dict': {
6004 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6005 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6006 'tags': [],
6007 },
6008 'playlist_mincount': 8,
6009 }, {
6010 # Should get three playlists for videos, shorts and streams tabs
6011 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6012 'info_dict': {
6013 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6014 'title': 'Polka Ch. 尾丸ポルカ',
6015 'channel_follower_count': int,
6016 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6017 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6018 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6019 'channel': 'Polka Ch. 尾丸ポルカ',
6020 'tags': 'count:35',
6021 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6022 'uploader': 'Polka Ch. 尾丸ポルカ',
6023 'uploader_id': '@OmaruPolka',
6024 },
6025 'playlist_count': 3,
6026 }, {
6027 # Shorts tab with channel with handle
6028 # TODO: fix channel description
6029 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6030 'info_dict': {
6031 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6032 'title': 'Not Just Bikes - Shorts',
6033 'tags': 'count:12',
6034 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6035 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6036 'channel_follower_count': int,
6037 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6038 'channel': 'Not Just Bikes',
6039 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6040 'uploader': 'Not Just Bikes',
6041 'uploader_id': '@NotJustBikes',
6042 },
6043 'playlist_mincount': 10,
6044 }, {
6045 # Streams tab
6046 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6047 'info_dict': {
6048 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6049 'title': '中村悠一 - Live',
6050 'tags': 'count:7',
6051 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6052 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6053 'channel': '中村悠一',
6054 'channel_follower_count': int,
6055 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6056 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6057 'uploader_id': '@Yuichi-Nakamura',
6058 'uploader': '中村悠一',
6059 },
6060 'playlist_mincount': 60,
6061 }, {
6062 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6063 # See test_youtube_lists
6064 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6065 'only_matching': True,
6066 }, {
6067 # No uploads and no UCID given. Should fail with no uploads error
6068 # See test_youtube_lists
6069 'url': 'https://www.youtube.com/news',
6070 'only_matching': True
6071 }, {
6072 # No videos tab but has a shorts tab
6073 'url': 'https://www.youtube.com/c/TKFShorts',
6074 'info_dict': {
6075 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6076 'title': 'Shorts Break - Shorts',
6077 'tags': 'count:48',
6078 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6079 'channel': 'Shorts Break',
6080 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6081 'channel_follower_count': int,
6082 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6083 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6084 'uploader': 'Shorts Break',
6085 'uploader_id': '@ShortsBreak_Official',
6086 },
6087 'playlist_mincount': 30,
6088 }, {
6089 # Trending Now Tab. tab id is empty
6090 'url': 'https://www.youtube.com/feed/trending',
6091 'info_dict': {
6092 'id': 'trending',
6093 'title': 'trending - Now',
6094 'tags': [],
6095 },
6096 'playlist_mincount': 30,
6097 }, {
6098 # Trending Gaming Tab. tab id is empty
6099 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6100 'info_dict': {
6101 'id': 'trending',
6102 'title': 'trending - Gaming',
6103 'tags': [],
6104 },
6105 'playlist_mincount': 30,
6106 }, {
6107 # Shorts url result in shorts tab
6108 # TODO: Fix channel id extraction
6109 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6110 'info_dict': {
6111 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6112 'title': 'cole-dlp-test-acc - Shorts',
6113 'channel': 'cole-dlp-test-acc',
6114 'description': 'test description',
6115 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6116 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6117 'tags': [],
6118 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6119 'uploader_id': '@coletdjnz',
6120 'uploader': 'cole-dlp-test-acc',
6121 },
6122 'playlist': [{
6123 'info_dict': {
6124 # Channel data is not currently available for short renderers (as of 2023-03-01)
6125 '_type': 'url',
6126 'ie_key': 'Youtube',
6127 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6128 'id': 'sSM9J5YH_60',
6129 'title': 'SHORT short',
6130 'view_count': int,
6131 'thumbnails': list,
6132 }
6133 }],
6134 'params': {'extract_flat': True},
6135 }, {
6136 # Live video status should be extracted
6137 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6138 'info_dict': {
6139 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6140 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6141 'tags': []
6142 },
6143 'playlist': [{
6144 'info_dict': {
6145 '_type': 'url',
6146 'ie_key': 'Youtube',
6147 'url': 'startswith:https://www.youtube.com/watch?v=',
6148 'id': str,
6149 'title': str,
6150 'live_status': 'is_live',
6151 'channel_id': str,
6152 'channel_url': str,
6153 'concurrent_view_count': int,
6154 'channel': str,
6155 }
6156 }],
6157 'params': {'extract_flat': True, 'playlist_items': '1'},
6158 'playlist_mincount': 1
6159 }, {
6160 # Channel renderer metadata. Contains number of videos on the channel
6161 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6162 'info_dict': {
6163 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6164 'title': 'cole-dlp-test-acc - Channels',
6165 'channel': 'cole-dlp-test-acc',
6166 'description': 'test description',
6167 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6168 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6169 'tags': [],
6170 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6171 'uploader_id': '@coletdjnz',
6172 'uploader': 'cole-dlp-test-acc',
6173 },
6174 'playlist': [{
6175 'info_dict': {
6176 '_type': 'url',
6177 'ie_key': 'YoutubeTab',
6178 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6179 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6180 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6181 'title': 'PewDiePie',
6182 'channel': 'PewDiePie',
6183 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6184 'thumbnails': list,
6185 'channel_follower_count': int,
6186 'playlist_count': int,
6187 'uploader': 'PewDiePie',
6188 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6189 'uploader_id': '@PewDiePie',
6190 }
6191 }],
6192 'params': {'extract_flat': True},
6193 }, {
6194 'url': 'https://www.youtube.com/@3blue1brown/about',
6195 'info_dict': {
6196 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6197 'tags': ['Mathematics'],
6198 'title': '3Blue1Brown - About',
6199 'channel_follower_count': int,
6200 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6201 'channel': '3Blue1Brown',
6202 'view_count': int,
6203 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6204 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6205 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6206 'uploader_id': '@3blue1brown',
6207 'uploader': '3Blue1Brown',
6208 },
6209 'playlist_count': 0,
6210 }, {
6211 # Podcasts tab, with rich entry playlistRenderers
6212 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6213 'info_dict': {
6214 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6215 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6216 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6217 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6218 'title': '99 Percent Invisible - Podcasts',
6219 'uploader': '99 Percent Invisible',
6220 'channel_follower_count': int,
6221 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6222 'tags': [],
6223 'channel': '99 Percent Invisible',
6224 'uploader_id': '@99percentinvisiblepodcast',
6225 },
6226 'playlist_count': 1,
6227 }, {
6228 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6229 'url': 'https://www.youtube.com/@AHimitsu/releases',
6230 'info_dict': {
6231 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6232 'channel': 'A Himitsu',
6233 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6234 'title': 'A Himitsu - Releases',
6235 'uploader_id': '@AHimitsu',
6236 'uploader': 'A Himitsu',
6237 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6238 'tags': 'count:16',
6239 'description': 'I make music',
6240 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6241 'channel_follower_count': int,
6242 },
6243 'playlist_mincount': 10,
6244 }]
6245
6246 @classmethod
6247 def suitable(cls, url):
6248 return False if YoutubeIE.suitable(url) else super().suitable(url)
6249
6250 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6251
6252 def _get_url_mobj(self, url):
6253 mobj = self._URL_RE.match(url).groupdict()
6254 mobj.update((k, '') for k, v in mobj.items() if v is None)
6255 return mobj
6256
6257 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6258 tab_name = (tab.get('title') or '').lower()
6259 tab_url = urljoin(base_url, traverse_obj(
6260 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6261
6262 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6263 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6264 if tab_id:
6265 return {
6266 'TAB_ID_SPONSORSHIPS': 'membership',
6267 }.get(tab_id, tab_id), tab_name
6268
6269 # Fallback to tab name if we cannot get the tab id.
6270 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6271 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6272 if tab_name:
6273 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6274 return {
6275 'home': 'featured',
6276 'live': 'streams',
6277 }.get(tab_name, tab_name), tab_name
6278
6279 def _has_tab(self, tabs, tab_id):
6280 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6281
6282 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6283 def _real_extract(self, url, smuggled_data):
6284 item_id = self._match_id(url)
6285 url = urllib.parse.urlunparse(
6286 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6287 compat_opts = self.get_param('compat_opts', [])
6288
6289 mobj = self._get_url_mobj(url)
6290 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6291 if is_channel and smuggled_data.get('is_music_url'):
6292 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6293 return self.url_result(
6294 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6295 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6296 mdata = self._extract_tab_endpoint(
6297 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6298 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6299 get_all=False, expected_type=str)
6300 if not murl:
6301 raise ExtractorError('Failed to resolve album to playlist')
6302 return self.url_result(murl, YoutubeTabIE)
6303 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6304 return self.url_result(
6305 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6306
6307 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6308 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6309 url = f'{pre}/videos{post}'
6310 if smuggled_data.get('is_music_url'):
6311 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6312
6313 # Handle both video/playlist URLs
6314 qs = parse_qs(url)
6315 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6316 if not video_id and mobj['not_channel'].startswith('watch'):
6317 if not playlist_id:
6318 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6319 raise ExtractorError('A video URL was given without video ID', expected=True)
6320 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6321 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6322 return self.url_result(
6323 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6324
6325 if not self._yes_playlist(playlist_id, video_id):
6326 return self.url_result(
6327 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6328
6329 data, ytcfg = self._extract_data(url, display_id)
6330
6331 # YouTube may provide a non-standard redirect to the regional channel
6332 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6333 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6334 redirect_url = traverse_obj(
6335 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6336 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6337 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6338 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6339 return self.url_result(redirect_url, YoutubeTabIE)
6340
6341 tabs, extra_tabs = self._extract_tab_renderers(data), []
6342 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6343 selected_tab = self._extract_selected_tab(tabs)
6344 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6345 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6346
6347 if not original_tab_id and selected_tab_name:
6348 self.to_screen('Downloading all uploads of the channel. '
6349 'To download only the videos in a specific tab, pass the tab\'s URL')
6350 if self._has_tab(tabs, 'streams'):
6351 extra_tabs.append(''.join((pre, '/streams', post)))
6352 if self._has_tab(tabs, 'shorts'):
6353 extra_tabs.append(''.join((pre, '/shorts', post)))
6354 # XXX: Members-only tab should also be extracted
6355
6356 if not extra_tabs and selected_tab_id != 'videos':
6357 # Channel does not have streams, shorts or videos tabs
6358 if item_id[:2] != 'UC':
6359 raise ExtractorError('This channel has no uploads', expected=True)
6360
6361 # Topic channels don't have /videos. Use the equivalent playlist instead
6362 pl_id = f'UU{item_id[2:]}'
6363 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6364 try:
6365 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6366 except ExtractorError:
6367 raise ExtractorError('This channel has no uploads', expected=True)
6368 else:
6369 item_id, url = pl_id, pl_url
6370 self.to_screen(
6371 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6372
6373 elif extra_tabs and selected_tab_id != 'videos':
6374 # When there are shorts/live tabs but not videos tab
6375 url, data = f'{pre}{post}', None
6376
6377 elif (original_tab_id or 'videos') != selected_tab_id:
6378 if original_tab_id == 'live':
6379 # Live tab should have redirected to the video
6380 # Except in the case the channel has an actual live tab
6381 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6382 raise UserNotLive(video_id=item_id)
6383 elif selected_tab_name:
6384 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6385
6386 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6387 url = f'{pre}{post}'
6388
6389 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6390 if 'no-youtube-unavailable-videos' not in compat_opts:
6391 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6392 self._extract_and_report_alerts(data, only_once=True)
6393
6394 tabs, entries = self._extract_tab_renderers(data), []
6395 if tabs:
6396 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6397 entries[0].update({
6398 'extractor_key': YoutubeTabIE.ie_key(),
6399 'extractor': YoutubeTabIE.IE_NAME,
6400 'webpage_url': url,
6401 })
6402 if self.get_param('playlist_items') == '0':
6403 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6404 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6405 entries.extend(map(self._real_extract, extra_tabs))
6406
6407 if len(entries) == 1:
6408 return entries[0]
6409 elif entries:
6410 metadata = self._extract_metadata_from_tabs(item_id, data)
6411 uploads_url = 'the Uploads (UU) playlist URL'
6412 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6413 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6414 self.to_screen(
6415 'Downloading as multiple playlists, separated by tabs. '
6416 f'To download as a single playlist instead, pass {uploads_url}')
6417 return self.playlist_result(entries, item_id, **metadata)
6418
6419 # Inline playlist
6420 playlist = traverse_obj(
6421 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6422 if playlist:
6423 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6424
6425 video_id = traverse_obj(
6426 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6427 if video_id:
6428 if tab != '/live': # live tab is expected to redirect to video
6429 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6430 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6431
6432 raise ExtractorError('Unable to recognize tab page')
6433
6434
6435 class YoutubePlaylistIE(InfoExtractor):
6436 IE_DESC = 'YouTube playlists'
6437 _VALID_URL = r'''(?x)(?:
6438 (?:https?://)?
6439 (?:\w+\.)?
6440 (?:
6441 (?:
6442 youtube(?:kids)?\.com|
6443 %(invidious)s
6444 )
6445 /.*?\?.*?\blist=
6446 )?
6447 (?P<id>%(playlist_id)s)
6448 )''' % {
6449 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6450 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6451 }
6452 IE_NAME = 'youtube:playlist'
6453 _TESTS = [{
6454 'note': 'issue #673',
6455 'url': 'PLBB231211A4F62143',
6456 'info_dict': {
6457 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6458 'id': 'PLBB231211A4F62143',
6459 'uploader': 'Wickman',
6460 'uploader_id': '@WickmanVT',
6461 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6462 'view_count': int,
6463 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6464 'modified_date': r're:\d{8}',
6465 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6466 'channel': 'Wickman',
6467 'tags': [],
6468 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6469 'availability': 'public',
6470 },
6471 'playlist_mincount': 29,
6472 }, {
6473 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6474 'info_dict': {
6475 'title': 'YDL_safe_search',
6476 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6477 },
6478 'playlist_count': 2,
6479 'skip': 'This playlist is private',
6480 }, {
6481 'note': 'embedded',
6482 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6483 'playlist_count': 4,
6484 'info_dict': {
6485 'title': 'JODA15',
6486 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6487 'uploader': 'milan',
6488 'uploader_id': '@milan5503',
6489 'description': '',
6490 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6491 'tags': [],
6492 'modified_date': '20140919',
6493 'view_count': int,
6494 'channel': 'milan',
6495 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6496 'uploader_url': 'https://www.youtube.com/@milan5503',
6497 'availability': 'public',
6498 },
6499 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6500 }, {
6501 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6502 'playlist_mincount': 455,
6503 'info_dict': {
6504 'title': '2018 Chinese New Singles (11/6 updated)',
6505 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6506 'uploader': 'LBK',
6507 'uploader_id': '@music_king',
6508 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6509 'channel': 'LBK',
6510 'view_count': int,
6511 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6512 'tags': [],
6513 'uploader_url': 'https://www.youtube.com/@music_king',
6514 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6515 'modified_date': r're:\d{8}',
6516 'availability': 'public',
6517 },
6518 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6519 }, {
6520 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6521 'only_matching': True,
6522 }, {
6523 # music album playlist
6524 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6525 'only_matching': True,
6526 }]
6527
6528 @classmethod
6529 def suitable(cls, url):
6530 if YoutubeTabIE.suitable(url):
6531 return False
6532 from ..utils import parse_qs
6533 qs = parse_qs(url)
6534 if qs.get('v', [None])[0]:
6535 return False
6536 return super().suitable(url)
6537
6538 def _real_extract(self, url):
6539 playlist_id = self._match_id(url)
6540 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6541 url = update_url_query(
6542 'https://www.youtube.com/playlist',
6543 parse_qs(url) or {'list': playlist_id})
6544 if is_music_url:
6545 url = smuggle_url(url, {'is_music_url': True})
6546 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6547
6548
6549 class YoutubeYtBeIE(InfoExtractor):
6550 IE_DESC = 'youtu.be'
6551 _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}
6552 _TESTS = [{
6553 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6554 'info_dict': {
6555 'id': 'yeWKywCrFtk',
6556 'ext': 'mp4',
6557 'title': 'Small Scale Baler and Braiding Rugs',
6558 'uploader': 'Backus-Page House Museum',
6559 'uploader_id': '@backuspagemuseum',
6560 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6561 'upload_date': '20161008',
6562 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6563 'categories': ['Nonprofits & Activism'],
6564 'tags': list,
6565 'like_count': int,
6566 'age_limit': 0,
6567 'playable_in_embed': True,
6568 'thumbnail': r're:^https?://.*\.webp',
6569 'channel': 'Backus-Page House Museum',
6570 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6571 'live_status': 'not_live',
6572 'view_count': int,
6573 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6574 'availability': 'public',
6575 'duration': 59,
6576 'comment_count': int,
6577 'channel_follower_count': int
6578 },
6579 'params': {
6580 'noplaylist': True,
6581 'skip_download': True,
6582 },
6583 }, {
6584 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6585 'only_matching': True,
6586 }]
6587
6588 def _real_extract(self, url):
6589 mobj = self._match_valid_url(url)
6590 video_id = mobj.group('id')
6591 playlist_id = mobj.group('playlist_id')
6592 return self.url_result(
6593 update_url_query('https://www.youtube.com/watch', {
6594 'v': video_id,
6595 'list': playlist_id,
6596 'feature': 'youtu.be',
6597 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6598
6599
6600 class YoutubeLivestreamEmbedIE(InfoExtractor):
6601 IE_DESC = 'YouTube livestream embeds'
6602 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6603 _TESTS = [{
6604 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6605 'only_matching': True,
6606 }]
6607
6608 def _real_extract(self, url):
6609 channel_id = self._match_id(url)
6610 return self.url_result(
6611 f'https://www.youtube.com/channel/{channel_id}/live',
6612 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6613
6614
6615 class YoutubeYtUserIE(InfoExtractor):
6616 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6617 IE_NAME = 'youtube:user'
6618 _VALID_URL = r'ytuser:(?P<id>.+)'
6619 _TESTS = [{
6620 'url': 'ytuser:phihag',
6621 'only_matching': True,
6622 }]
6623
6624 def _real_extract(self, url):
6625 user_id = self._match_id(url)
6626 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6627
6628
6629 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6630 IE_NAME = 'youtube:favorites'
6631 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6632 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6633 _LOGIN_REQUIRED = True
6634 _TESTS = [{
6635 'url': ':ytfav',
6636 'only_matching': True,
6637 }, {
6638 'url': ':ytfavorites',
6639 'only_matching': True,
6640 }]
6641
6642 def _real_extract(self, url):
6643 return self.url_result(
6644 'https://www.youtube.com/playlist?list=LL',
6645 ie=YoutubeTabIE.ie_key())
6646
6647
6648 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6649 IE_NAME = 'youtube:notif'
6650 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6651 _VALID_URL = r':ytnotif(?:ication)?s?'
6652 _LOGIN_REQUIRED = True
6653 _TESTS = [{
6654 'url': ':ytnotif',
6655 'only_matching': True,
6656 }, {
6657 'url': ':ytnotifications',
6658 'only_matching': True,
6659 }]
6660
6661 def _extract_notification_menu(self, response, continuation_list):
6662 notification_list = traverse_obj(
6663 response,
6664 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6665 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6666 expected_type=list) or []
6667 continuation_list[0] = None
6668 for item in notification_list:
6669 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6670 if entry:
6671 yield entry
6672 continuation = item.get('continuationItemRenderer')
6673 if continuation:
6674 continuation_list[0] = continuation
6675
6676 def _extract_notification_renderer(self, notification):
6677 video_id = traverse_obj(
6678 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6679 url = f'https://www.youtube.com/watch?v={video_id}'
6680 channel_id = None
6681 if not video_id:
6682 browse_ep = traverse_obj(
6683 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6684 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6685 post_id = self._search_regex(
6686 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6687 'post id', default=None)
6688 if not channel_id or not post_id:
6689 return
6690 # The direct /post url redirects to this in the browser
6691 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6692
6693 channel = traverse_obj(
6694 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6695 expected_type=str)
6696 notification_title = self._get_text(notification, 'shortMessage')
6697 if notification_title:
6698 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6699 # TODO: handle recommended videos
6700 title = self._search_regex(
6701 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6702 'video title', default=None)
6703 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6704 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6705 else None)
6706 return {
6707 '_type': 'url',
6708 'url': url,
6709 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6710 'video_id': video_id,
6711 'title': title,
6712 'channel_id': channel_id,
6713 'channel': channel,
6714 'uploader': channel,
6715 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6716 'timestamp': timestamp,
6717 }
6718
6719 def _notification_menu_entries(self, ytcfg):
6720 continuation_list = [None]
6721 response = None
6722 for page in itertools.count(1):
6723 ctoken = traverse_obj(
6724 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6725 response = self._extract_response(
6726 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6727 ep='notification/get_notification_menu', check_get_keys='actions',
6728 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6729 yield from self._extract_notification_menu(response, continuation_list)
6730 if not continuation_list[0]:
6731 break
6732
6733 def _real_extract(self, url):
6734 display_id = 'notifications'
6735 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6736 self._report_playlist_authcheck(ytcfg)
6737 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6738
6739
6740 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6741 IE_DESC = 'YouTube search'
6742 IE_NAME = 'youtube:search'
6743 _SEARCH_KEY = 'ytsearch'
6744 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6745 _TESTS = [{
6746 'url': 'ytsearch5:youtube-dl test video',
6747 'playlist_count': 5,
6748 'info_dict': {
6749 'id': 'youtube-dl test video',
6750 'title': 'youtube-dl test video',
6751 }
6752 }]
6753
6754
6755 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6756 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6757 _SEARCH_KEY = 'ytsearchdate'
6758 IE_DESC = 'YouTube search, newest videos first'
6759 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6760 _TESTS = [{
6761 'url': 'ytsearchdate5:youtube-dl test video',
6762 'playlist_count': 5,
6763 'info_dict': {
6764 'id': 'youtube-dl test video',
6765 'title': 'youtube-dl test video',
6766 }
6767 }]
6768
6769
6770 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6771 IE_DESC = 'YouTube search URLs with sorting and filter support'
6772 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6773 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6774 _TESTS = [{
6775 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6776 'playlist_mincount': 5,
6777 'info_dict': {
6778 'id': 'youtube-dl test video',
6779 'title': 'youtube-dl test video',
6780 }
6781 }, {
6782 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6783 'playlist_mincount': 5,
6784 'info_dict': {
6785 'id': 'python',
6786 'title': 'python',
6787 }
6788 }, {
6789 'url': 'https://www.youtube.com/results?search_query=%23cats',
6790 'playlist_mincount': 1,
6791 'info_dict': {
6792 'id': '#cats',
6793 'title': '#cats',
6794 # The test suite does not have support for nested playlists
6795 # 'entries': [{
6796 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6797 # 'title': '#cats',
6798 # }],
6799 },
6800 }, {
6801 # Channel results
6802 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6803 'info_dict': {
6804 'id': 'kurzgesagt',
6805 'title': 'kurzgesagt',
6806 },
6807 'playlist': [{
6808 'info_dict': {
6809 '_type': 'url',
6810 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6811 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6812 'ie_key': 'YoutubeTab',
6813 'channel': 'Kurzgesagt – In a Nutshell',
6814 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6815 'title': 'Kurzgesagt – In a Nutshell',
6816 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6817 'playlist_count': int, # XXX: should have a way of saying > 1
6818 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6819 'thumbnails': list,
6820 'uploader_id': '@kurzgesagt',
6821 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
6822 'uploader': 'Kurzgesagt – In a Nutshell',
6823 }
6824 }],
6825 'params': {'extract_flat': True, 'playlist_items': '1'},
6826 'playlist_mincount': 1,
6827 }, {
6828 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
6829 'only_matching': True,
6830 }]
6831
6832 def _real_extract(self, url):
6833 qs = parse_qs(url)
6834 query = (qs.get('search_query') or qs.get('q'))[0]
6835 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
6836
6837
6838 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
6839 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
6840 IE_NAME = 'youtube:music:search_url'
6841 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6842 _TESTS = [{
6843 'url': 'https://music.youtube.com/search?q=royalty+free+music',
6844 'playlist_count': 16,
6845 'info_dict': {
6846 'id': 'royalty free music',
6847 'title': 'royalty free music',
6848 }
6849 }, {
6850 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
6851 'playlist_mincount': 30,
6852 'info_dict': {
6853 'id': 'royalty free music - songs',
6854 'title': 'royalty free music - songs',
6855 },
6856 'params': {'extract_flat': 'in_playlist'}
6857 }, {
6858 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
6859 'playlist_mincount': 30,
6860 'info_dict': {
6861 'id': 'royalty free music - community playlists',
6862 'title': 'royalty free music - community playlists',
6863 },
6864 'params': {'extract_flat': 'in_playlist'}
6865 }]
6866
6867 _SECTIONS = {
6868 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
6869 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
6870 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
6871 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
6872 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
6873 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
6874 }
6875
6876 def _real_extract(self, url):
6877 qs = parse_qs(url)
6878 query = (qs.get('search_query') or qs.get('q'))[0]
6879 params = qs.get('sp', (None,))[0]
6880 if params:
6881 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
6882 else:
6883 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
6884 params = self._SECTIONS.get(section)
6885 if not params:
6886 section = None
6887 title = join_nonempty(query, section, delim=' - ')
6888 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
6889
6890
6891 class YoutubeFeedsInfoExtractor(InfoExtractor):
6892 """
6893 Base class for feed extractors
6894 Subclasses must re-define the _FEED_NAME property.
6895 """
6896 _LOGIN_REQUIRED = True
6897 _FEED_NAME = 'feeds'
6898
6899 def _real_initialize(self):
6900 YoutubeBaseInfoExtractor._check_login_required(self)
6901
6902 @classproperty
6903 def IE_NAME(self):
6904 return f'youtube:{self._FEED_NAME}'
6905
6906 def _real_extract(self, url):
6907 return self.url_result(
6908 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
6909
6910
6911 class YoutubeWatchLaterIE(InfoExtractor):
6912 IE_NAME = 'youtube:watchlater'
6913 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
6914 _VALID_URL = r':ytwatchlater'
6915 _TESTS = [{
6916 'url': ':ytwatchlater',
6917 'only_matching': True,
6918 }]
6919
6920 def _real_extract(self, url):
6921 return self.url_result(
6922 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
6923
6924
6925 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
6926 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
6927 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
6928 _FEED_NAME = 'recommended'
6929 _LOGIN_REQUIRED = False
6930 _TESTS = [{
6931 'url': ':ytrec',
6932 'only_matching': True,
6933 }, {
6934 'url': ':ytrecommended',
6935 'only_matching': True,
6936 }, {
6937 'url': 'https://youtube.com',
6938 'only_matching': True,
6939 }]
6940
6941
6942 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
6943 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
6944 _VALID_URL = r':ytsub(?:scription)?s?'
6945 _FEED_NAME = 'subscriptions'
6946 _TESTS = [{
6947 'url': ':ytsubs',
6948 'only_matching': True,
6949 }, {
6950 'url': ':ytsubscriptions',
6951 'only_matching': True,
6952 }]
6953
6954
6955 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
6956 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
6957 _VALID_URL = r':ythis(?:tory)?'
6958 _FEED_NAME = 'history'
6959 _TESTS = [{
6960 'url': ':ythistory',
6961 'only_matching': True,
6962 }]
6963
6964
6965 class YoutubeStoriesIE(InfoExtractor):
6966 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
6967 IE_NAME = 'youtube:stories'
6968 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
6969 _TESTS = [{
6970 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
6971 'only_matching': True,
6972 }]
6973
6974 def _real_extract(self, url):
6975 playlist_id = f'RLTD{self._match_id(url)}'
6976 return self.url_result(
6977 smuggle_url(f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1', {'is_story': True}),
6978 ie=YoutubeTabIE, video_id=playlist_id)
6979
6980
6981 class YoutubeShortsAudioPivotIE(InfoExtractor):
6982 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
6983 IE_NAME = 'youtube:shorts:pivot:audio'
6984 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
6985 _TESTS = [{
6986 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
6987 'only_matching': True,
6988 }]
6989
6990 @staticmethod
6991 def _generate_audio_pivot_params(video_id):
6992 """
6993 Generates sfv_audio_pivot browse params for this video id
6994 """
6995 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
6996 return urllib.parse.quote(base64.b64encode(pb_params).decode())
6997
6998 def _real_extract(self, url):
6999 video_id = self._match_id(url)
7000 return self.url_result(
7001 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7002 ie=YoutubeTabIE)
7003
7004
7005 class YoutubeTruncatedURLIE(InfoExtractor):
7006 IE_NAME = 'youtube:truncated_url'
7007 IE_DESC = False # Do not list
7008 _VALID_URL = r'''(?x)
7009 (?:https?://)?
7010 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7011 (?:watch\?(?:
7012 feature=[a-z_]+|
7013 annotation_id=annotation_[^&]+|
7014 x-yt-cl=[0-9]+|
7015 hl=[^&]*|
7016 t=[0-9]+
7017 )?
7018 |
7019 attribution_link\?a=[^&]+
7020 )
7021 $
7022 '''
7023
7024 _TESTS = [{
7025 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7026 'only_matching': True,
7027 }, {
7028 'url': 'https://www.youtube.com/watch?',
7029 'only_matching': True,
7030 }, {
7031 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7032 'only_matching': True,
7033 }, {
7034 'url': 'https://www.youtube.com/watch?feature=foo',
7035 'only_matching': True,
7036 }, {
7037 'url': 'https://www.youtube.com/watch?hl=en-GB',
7038 'only_matching': True,
7039 }, {
7040 'url': 'https://www.youtube.com/watch?t=2372',
7041 'only_matching': True,
7042 }]
7043
7044 def _real_extract(self, url):
7045 raise ExtractorError(
7046 'Did you forget to quote the URL? Remember that & is a meta '
7047 'character in most shells, so you want to put the URL in quotes, '
7048 'like youtube-dl '
7049 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7050 ' or simply youtube-dl BaW_jenozKc .',
7051 expected=True)
7052
7053
7054 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7055 IE_NAME = 'youtube:clip'
7056 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7057 _TESTS = [{
7058 # FIXME: Other metadata should be extracted from the clip, not from the base video
7059 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7060 'info_dict': {
7061 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7062 'ext': 'mp4',
7063 'section_start': 29.0,
7064 'section_end': 39.7,
7065 'duration': 10.7,
7066 'age_limit': 0,
7067 'availability': 'public',
7068 'categories': ['Gaming'],
7069 'channel': 'Scott The Woz',
7070 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7071 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7072 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7073 'like_count': int,
7074 'playable_in_embed': True,
7075 'tags': 'count:17',
7076 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7077 'title': 'Mobile Games on Console - Scott The Woz',
7078 'upload_date': '20210920',
7079 'uploader': 'Scott The Woz',
7080 'uploader_id': '@ScottTheWoz',
7081 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7082 'view_count': int,
7083 'live_status': 'not_live',
7084 'channel_follower_count': int,
7085 'chapters': 'count:20',
7086 }
7087 }]
7088
7089 def _real_extract(self, url):
7090 clip_id = self._match_id(url)
7091 _, data = self._extract_webpage(url, clip_id)
7092
7093 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7094 if not video_id:
7095 raise ExtractorError('Unable to find video ID')
7096
7097 clip_data = traverse_obj(data, (
7098 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7099 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7100 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7101 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7102
7103 return {
7104 '_type': 'url_transparent',
7105 'url': f'https://www.youtube.com/watch?v={video_id}',
7106 'ie_key': YoutubeIE.ie_key(),
7107 'id': clip_id,
7108 'section_start': int(clip_data['startTimeMs']) / 1000,
7109 'section_end': int(clip_data['endTimeMs']) / 1000,
7110 }
7111
7112
7113 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7114 IE_NAME = 'youtube:consent'
7115 IE_DESC = False # Do not list
7116 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7117 _TESTS = [{
7118 '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',
7119 'info_dict': {
7120 'id': 'qVv6vCqciTM',
7121 'ext': 'mp4',
7122 'age_limit': 0,
7123 'uploader_id': '@sana_natori',
7124 'comment_count': int,
7125 'chapters': 'count:13',
7126 'upload_date': '20221223',
7127 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7128 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7129 'uploader_url': 'https://www.youtube.com/@sana_natori',
7130 'like_count': int,
7131 'release_date': '20221223',
7132 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7133 '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】',
7134 'view_count': int,
7135 'playable_in_embed': True,
7136 'duration': 4438,
7137 'availability': 'public',
7138 'channel_follower_count': int,
7139 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7140 'categories': ['Entertainment'],
7141 'live_status': 'was_live',
7142 'release_timestamp': 1671793345,
7143 'channel': 'さなちゃんねる',
7144 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7145 'uploader': 'さなちゃんねる',
7146 },
7147 'add_ie': ['Youtube'],
7148 'params': {'skip_download': 'Youtube'},
7149 }]
7150
7151 def _real_extract(self, url):
7152 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7153 if not redirect_url:
7154 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7155 return self.url_result(redirect_url)
7156
7157
7158 class YoutubeTruncatedIDIE(InfoExtractor):
7159 IE_NAME = 'youtube:truncated_id'
7160 IE_DESC = False # Do not list
7161 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7162
7163 _TESTS = [{
7164 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7165 'only_matching': True,
7166 }]
7167
7168 def _real_extract(self, url):
7169 video_id = self._match_id(url)
7170 raise ExtractorError(
7171 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7172 expected=True)