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