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