]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[extractor/youtube] Process `post_live` over 2 hours
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime
6 import enum
7 import hashlib
8 import itertools
9 import json
10 import math
11 import os.path
12 import random
13 import re
14 import sys
15 import threading
16 import time
17 import traceback
18 import urllib.error
19 import urllib.parse
20
21 from .common import InfoExtractor, SearchInfoExtractor
22 from .openload import PhantomJSwrapper
23 from ..compat import functools
24 from ..jsinterp import JSInterpreter
25 from ..utils import (
26 NO_DEFAULT,
27 ExtractorError,
28 LazyList,
29 UserNotLive,
30 bug_reports_message,
31 classproperty,
32 clean_html,
33 datetime_from_str,
34 dict_get,
35 filter_dict,
36 float_or_none,
37 format_field,
38 get_first,
39 int_or_none,
40 is_html,
41 join_nonempty,
42 js_to_json,
43 mimetype2ext,
44 network_exceptions,
45 orderedSet,
46 parse_codecs,
47 parse_count,
48 parse_duration,
49 parse_iso8601,
50 parse_qs,
51 qualities,
52 remove_start,
53 smuggle_url,
54 str_or_none,
55 str_to_int,
56 strftime_or_none,
57 traverse_obj,
58 try_get,
59 unescapeHTML,
60 unified_strdate,
61 unified_timestamp,
62 unsmuggle_url,
63 update_url_query,
64 url_or_none,
65 urljoin,
66 variadic,
67 )
68
69 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
70 # any clients starting with _ cannot be explicitly requested by the user
71 INNERTUBE_CLIENTS = {
72 'web': {
73 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
74 'INNERTUBE_CONTEXT': {
75 'client': {
76 'clientName': 'WEB',
77 'clientVersion': '2.20220801.00.00',
78 }
79 },
80 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
81 },
82 'web_embedded': {
83 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
84 'INNERTUBE_CONTEXT': {
85 'client': {
86 'clientName': 'WEB_EMBEDDED_PLAYER',
87 'clientVersion': '1.20220731.00.00',
88 },
89 },
90 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
91 },
92 'web_music': {
93 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
94 'INNERTUBE_HOST': 'music.youtube.com',
95 'INNERTUBE_CONTEXT': {
96 'client': {
97 'clientName': 'WEB_REMIX',
98 'clientVersion': '1.20220727.01.00',
99 }
100 },
101 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
102 },
103 'web_creator': {
104 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
105 'INNERTUBE_CONTEXT': {
106 'client': {
107 'clientName': 'WEB_CREATOR',
108 'clientVersion': '1.20220726.00.00',
109 }
110 },
111 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
112 },
113 'android': {
114 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
115 'INNERTUBE_CONTEXT': {
116 'client': {
117 'clientName': 'ANDROID',
118 'clientVersion': '17.31.35',
119 'androidSdkVersion': 30,
120 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
121 }
122 },
123 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
124 'REQUIRE_JS_PLAYER': False
125 },
126 'android_embedded': {
127 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
128 'INNERTUBE_CONTEXT': {
129 'client': {
130 'clientName': 'ANDROID_EMBEDDED_PLAYER',
131 'clientVersion': '17.31.35',
132 'androidSdkVersion': 30,
133 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
134 },
135 },
136 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
137 'REQUIRE_JS_PLAYER': False
138 },
139 'android_music': {
140 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
141 'INNERTUBE_CONTEXT': {
142 'client': {
143 'clientName': 'ANDROID_MUSIC',
144 'clientVersion': '5.16.51',
145 'androidSdkVersion': 30,
146 'userAgent': 'com.google.android.apps.youtube.music/5.16.51 (Linux; U; Android 11) gzip'
147 }
148 },
149 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
150 'REQUIRE_JS_PLAYER': False
151 },
152 'android_creator': {
153 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
154 'INNERTUBE_CONTEXT': {
155 'client': {
156 'clientName': 'ANDROID_CREATOR',
157 'clientVersion': '22.30.100',
158 'androidSdkVersion': 30,
159 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
160 },
161 },
162 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
163 'REQUIRE_JS_PLAYER': False
164 },
165 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
166 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
167 'ios': {
168 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
169 'INNERTUBE_CONTEXT': {
170 'client': {
171 'clientName': 'IOS',
172 'clientVersion': '17.33.2',
173 'deviceModel': 'iPhone14,3',
174 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
175 }
176 },
177 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
178 'REQUIRE_JS_PLAYER': False
179 },
180 'ios_embedded': {
181 'INNERTUBE_CONTEXT': {
182 'client': {
183 'clientName': 'IOS_MESSAGES_EXTENSION',
184 'clientVersion': '17.33.2',
185 'deviceModel': 'iPhone14,3',
186 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
187 },
188 },
189 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
190 'REQUIRE_JS_PLAYER': False
191 },
192 'ios_music': {
193 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
194 'INNERTUBE_CONTEXT': {
195 'client': {
196 'clientName': 'IOS_MUSIC',
197 'clientVersion': '5.21',
198 'deviceModel': 'iPhone14,3',
199 'userAgent': 'com.google.ios.youtubemusic/5.21 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
200 },
201 },
202 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
203 'REQUIRE_JS_PLAYER': False
204 },
205 'ios_creator': {
206 'INNERTUBE_CONTEXT': {
207 'client': {
208 'clientName': 'IOS_CREATOR',
209 'clientVersion': '22.33.101',
210 'deviceModel': 'iPhone14,3',
211 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
212 },
213 },
214 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
215 'REQUIRE_JS_PLAYER': False
216 },
217 # mweb has 'ultralow' formats
218 # See: https://github.com/yt-dlp/yt-dlp/pull/557
219 'mweb': {
220 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
221 'INNERTUBE_CONTEXT': {
222 'client': {
223 'clientName': 'MWEB',
224 'clientVersion': '2.20220801.00.00',
225 }
226 },
227 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
228 },
229 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
230 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
231 'tv_embedded': {
232 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
233 'INNERTUBE_CONTEXT': {
234 'client': {
235 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
236 'clientVersion': '2.0',
237 },
238 },
239 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
240 },
241 }
242
243
244 def _split_innertube_client(client_name):
245 variant, *base = client_name.rsplit('.', 1)
246 if base:
247 return variant, base[0], variant
248 base, *variant = client_name.split('_', 1)
249 return client_name, base, variant[0] if variant else None
250
251
252 def short_client_name(client_name):
253 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
254 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
255
256
257 def build_innertube_clients():
258 THIRD_PARTY = {
259 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
260 }
261 BASE_CLIENTS = ('ios', 'android', 'web', 'tv', 'mweb')
262 priority = qualities(BASE_CLIENTS[::-1])
263
264 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
265 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
266 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
267 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
268 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
269
270 _, base_client, variant = _split_innertube_client(client)
271 ytcfg['priority'] = 10 * priority(base_client)
272
273 if not variant:
274 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
275 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
276 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
277 embedscreen['priority'] -= 3
278 elif variant == 'embedded':
279 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
280 ytcfg['priority'] -= 2
281 else:
282 ytcfg['priority'] -= 3
283
284
285 build_innertube_clients()
286
287
288 class BadgeType(enum.Enum):
289 AVAILABILITY_UNLISTED = enum.auto()
290 AVAILABILITY_PRIVATE = enum.auto()
291 AVAILABILITY_PUBLIC = enum.auto()
292 AVAILABILITY_PREMIUM = enum.auto()
293 AVAILABILITY_SUBSCRIPTION = enum.auto()
294 LIVE_NOW = enum.auto()
295 VERIFIED = enum.auto()
296
297
298 class YoutubeBaseInfoExtractor(InfoExtractor):
299 """Provide base functions for Youtube extractors"""
300
301 _RESERVED_NAMES = (
302 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
303 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
304 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
305 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
306
307 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
308
309 # _NETRC_MACHINE = 'youtube'
310
311 # If True it will raise an error if no login info is provided
312 _LOGIN_REQUIRED = False
313
314 _INVIDIOUS_SITES = (
315 # invidious-redirect websites
316 r'(?:www\.)?redirect\.invidious\.io',
317 r'(?:(?:www|dev)\.)?invidio\.us',
318 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
319 r'(?:www\.)?invidious\.pussthecat\.org',
320 r'(?:www\.)?invidious\.zee\.li',
321 r'(?:www\.)?invidious\.ethibox\.fr',
322 r'(?:www\.)?iv\.ggtyler\.dev',
323 r'(?:www\.)?inv\.vern\.i2p',
324 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
325 r'(?:www\.)?inv\.riverside\.rocks',
326 r'(?:www\.)?invidious\.silur\.me',
327 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
328 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
329 r'(?:www\.)?invidious\.slipfox\.xyz',
330 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
331 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
332 r'(?:www\.)?invidious\.tiekoetter\.com',
333 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
334 r'(?:www\.)?invidious\.nerdvpn\.de',
335 r'(?:www\.)?invidious\.weblibre\.org',
336 r'(?:www\.)?inv\.odyssey346\.dev',
337 r'(?:www\.)?invidious\.dhusch\.de',
338 r'(?:www\.)?iv\.melmac\.space',
339 r'(?:www\.)?watch\.thekitty\.zone',
340 r'(?:www\.)?invidious\.privacydev\.net',
341 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
342 r'(?:www\.)?invidious\.drivet\.xyz',
343 r'(?:www\.)?vid\.priv\.au',
344 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
345 r'(?:www\.)?inv\.vern\.cc',
346 r'(?:www\.)?invidious\.esmailelbob\.xyz',
347 r'(?:www\.)?invidious\.sethforprivacy\.com',
348 r'(?:www\.)?yt\.oelrichsgarcia\.de',
349 r'(?:www\.)?yt\.artemislena\.eu',
350 r'(?:www\.)?invidious\.flokinet\.to',
351 r'(?:www\.)?invidious\.baczek\.me',
352 r'(?:www\.)?y\.com\.sb',
353 r'(?:www\.)?invidious\.epicsite\.xyz',
354 r'(?:www\.)?invidious\.lidarshield\.cloud',
355 r'(?:www\.)?yt\.funami\.tech',
356 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
357 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
358 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
359 # youtube-dl invidious instances list
360 r'(?:(?:www|no)\.)?invidiou\.sh',
361 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
362 r'(?:www\.)?invidious\.kabi\.tk',
363 r'(?:www\.)?invidious\.mastodon\.host',
364 r'(?:www\.)?invidious\.zapashcanon\.fr',
365 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
366 r'(?:www\.)?invidious\.tinfoil-hat\.net',
367 r'(?:www\.)?invidious\.himiko\.cloud',
368 r'(?:www\.)?invidious\.reallyancient\.tech',
369 r'(?:www\.)?invidious\.tube',
370 r'(?:www\.)?invidiou\.site',
371 r'(?:www\.)?invidious\.site',
372 r'(?:www\.)?invidious\.xyz',
373 r'(?:www\.)?invidious\.nixnet\.xyz',
374 r'(?:www\.)?invidious\.048596\.xyz',
375 r'(?:www\.)?invidious\.drycat\.fr',
376 r'(?:www\.)?inv\.skyn3t\.in',
377 r'(?:www\.)?tube\.poal\.co',
378 r'(?:www\.)?tube\.connect\.cafe',
379 r'(?:www\.)?vid\.wxzm\.sx',
380 r'(?:www\.)?vid\.mint\.lgbt',
381 r'(?:www\.)?vid\.puffyan\.us',
382 r'(?:www\.)?yewtu\.be',
383 r'(?:www\.)?yt\.elukerio\.org',
384 r'(?:www\.)?yt\.lelux\.fi',
385 r'(?:www\.)?invidious\.ggc-project\.de',
386 r'(?:www\.)?yt\.maisputain\.ovh',
387 r'(?:www\.)?ytprivate\.com',
388 r'(?:www\.)?invidious\.13ad\.de',
389 r'(?:www\.)?invidious\.toot\.koeln',
390 r'(?:www\.)?invidious\.fdn\.fr',
391 r'(?:www\.)?watch\.nettohikari\.com',
392 r'(?:www\.)?invidious\.namazso\.eu',
393 r'(?:www\.)?invidious\.silkky\.cloud',
394 r'(?:www\.)?invidious\.exonip\.de',
395 r'(?:www\.)?invidious\.riverside\.rocks',
396 r'(?:www\.)?invidious\.blamefran\.net',
397 r'(?:www\.)?invidious\.moomoo\.de',
398 r'(?:www\.)?ytb\.trom\.tf',
399 r'(?:www\.)?yt\.cyberhost\.uk',
400 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
401 r'(?:www\.)?qklhadlycap4cnod\.onion',
402 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
403 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
404 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
405 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
406 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
407 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
408 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
409 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
410 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
411 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
412 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
413 r'(?:www\.)?piped\.kavin\.rocks',
414 r'(?:www\.)?piped\.tokhmi\.xyz',
415 r'(?:www\.)?piped\.syncpundit\.io',
416 r'(?:www\.)?piped\.mha\.fi',
417 r'(?:www\.)?watch\.whatever\.social',
418 r'(?:www\.)?piped\.garudalinux\.org',
419 r'(?:www\.)?piped\.rivo\.lol',
420 r'(?:www\.)?piped-libre\.kavin\.rocks',
421 r'(?:www\.)?yt\.jae\.fi',
422 r'(?:www\.)?piped\.mint\.lgbt',
423 r'(?:www\.)?il\.ax',
424 r'(?:www\.)?piped\.esmailelbob\.xyz',
425 r'(?:www\.)?piped\.projectsegfau\.lt',
426 r'(?:www\.)?piped\.privacydev\.net',
427 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
428 r'(?:www\.)?piped\.smnz\.de',
429 r'(?:www\.)?piped\.adminforge\.de',
430 r'(?:www\.)?watch\.whatevertinfoil\.de',
431 r'(?:www\.)?piped\.qdi\.fi',
432 r'(?:www\.)?piped\.video',
433 r'(?:www\.)?piped\.aeong\.one',
434 r'(?:www\.)?piped\.moomoo\.me',
435 r'(?:www\.)?piped\.chauvet\.pro',
436 r'(?:www\.)?watch\.leptons\.xyz',
437 r'(?:www\.)?pd\.vern\.cc',
438 r'(?:www\.)?piped\.hostux\.net',
439 r'(?:www\.)?piped\.lunar\.icu',
440 # Hyperpipe instances from https://hyperpipe.codeberg.page/
441 r'(?:www\.)?hyperpipe\.surge\.sh',
442 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
443 r'(?:www\.)?listen\.whatever\.social',
444 r'(?:www\.)?music\.adminforge\.de',
445 )
446
447 # extracted from account/account_menu ep
448 # XXX: These are the supported YouTube UI and API languages,
449 # which is slightly different from languages supported for translation in YouTube studio
450 _SUPPORTED_LANG_CODES = [
451 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
452 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
453 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
454 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
455 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
456 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
457 ]
458
459 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
460
461 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
462 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
463
464 def ucid_or_none(self, ucid):
465 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
466
467 def handle_or_none(self, handle):
468 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
469
470 def handle_from_url(self, url):
471 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
472 url, 'channel handle', default=None)
473
474 def ucid_from_url(self, url):
475 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
476 url, 'channel id', default=None)
477
478 @functools.cached_property
479 def _preferred_lang(self):
480 """
481 Returns a language code supported by YouTube for the user preferred language.
482 Returns None if no preferred language set.
483 """
484 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
485 if not preferred_lang:
486 return
487 if preferred_lang not in self._SUPPORTED_LANG_CODES:
488 raise ExtractorError(
489 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
490 expected=True)
491 elif preferred_lang != 'en':
492 self.report_warning(
493 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
494 return preferred_lang
495
496 def _initialize_consent(self):
497 cookies = self._get_cookies('https://www.youtube.com/')
498 if cookies.get('__Secure-3PSID'):
499 return
500 consent_id = None
501 consent = cookies.get('CONSENT')
502 if consent:
503 if 'YES' in consent.value:
504 return
505 consent_id = self._search_regex(
506 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
507 if not consent_id:
508 consent_id = random.randint(100, 999)
509 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
510
511 def _initialize_pref(self):
512 cookies = self._get_cookies('https://www.youtube.com/')
513 pref_cookie = cookies.get('PREF')
514 pref = {}
515 if pref_cookie:
516 try:
517 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
518 except ValueError:
519 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
520 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
521 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
522
523 def _real_initialize(self):
524 self._initialize_pref()
525 self._initialize_consent()
526 self._check_login_required()
527
528 def _check_login_required(self):
529 if self._LOGIN_REQUIRED and not self._cookies_passed:
530 self.raise_login_required('Login details are needed to download this content', method='cookies')
531
532 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
533 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
534
535 def _get_default_ytcfg(self, client='web'):
536 return copy.deepcopy(INNERTUBE_CLIENTS[client])
537
538 def _get_innertube_host(self, client='web'):
539 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
540
541 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
542 # try_get but with fallback to default ytcfg client values when present
543 _func = lambda y: try_get(y, getter, expected_type)
544 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
545
546 def _extract_client_name(self, ytcfg, default_client='web'):
547 return self._ytcfg_get_safe(
548 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
549 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
550
551 def _extract_client_version(self, ytcfg, default_client='web'):
552 return self._ytcfg_get_safe(
553 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
554 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
555
556 def _select_api_hostname(self, req_api_hostname, default_client=None):
557 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
558 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
559
560 def _extract_api_key(self, ytcfg=None, default_client='web'):
561 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
562
563 def _extract_context(self, ytcfg=None, default_client='web'):
564 context = get_first(
565 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
566 # Enforce language and tz for extraction
567 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
568 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
569 return context
570
571 _SAPISID = None
572
573 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
574 time_now = round(time.time())
575 if self._SAPISID is None:
576 yt_cookies = self._get_cookies('https://www.youtube.com')
577 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
578 # See: https://github.com/yt-dlp/yt-dlp/issues/393
579 sapisid_cookie = dict_get(
580 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
581 if sapisid_cookie and sapisid_cookie.value:
582 self._SAPISID = sapisid_cookie.value
583 self.write_debug('Extracted SAPISID cookie')
584 # SAPISID cookie is required if not already present
585 if not yt_cookies.get('SAPISID'):
586 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
587 self._set_cookie(
588 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
589 else:
590 self._SAPISID = False
591 if not self._SAPISID:
592 return None
593 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
594 sapisidhash = hashlib.sha1(
595 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
596 return f'SAPISIDHASH {time_now}_{sapisidhash}'
597
598 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
599 note='Downloading API JSON', errnote='Unable to download API page',
600 context=None, api_key=None, api_hostname=None, default_client='web'):
601
602 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
603 data.update(query)
604 real_headers = self.generate_api_headers(default_client=default_client)
605 real_headers.update({'content-type': 'application/json'})
606 if headers:
607 real_headers.update(headers)
608 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
609 or api_key or self._extract_api_key(default_client=default_client))
610 return self._download_json(
611 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
612 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
613 data=json.dumps(data).encode('utf8'), headers=real_headers,
614 query={'key': api_key, 'prettyPrint': 'false'})
615
616 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
617 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
618
619 @staticmethod
620 def _extract_session_index(*data):
621 """
622 Index of current account in account list.
623 See: https://github.com/yt-dlp/yt-dlp/pull/519
624 """
625 for ytcfg in data:
626 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
627 if session_index is not None:
628 return session_index
629
630 # Deprecated?
631 def _extract_identity_token(self, ytcfg=None, webpage=None):
632 if ytcfg:
633 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
634 if token:
635 return token
636 if webpage:
637 return self._search_regex(
638 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
639 'identity token', default=None, fatal=False)
640
641 @staticmethod
642 def _extract_account_syncid(*args):
643 """
644 Extract syncId required to download private playlists of secondary channels
645 @params response and/or ytcfg
646 """
647 for data in args:
648 # ytcfg includes channel_syncid if on secondary channel
649 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
650 if delegated_sid:
651 return delegated_sid
652 sync_ids = (try_get(
653 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
654 lambda x: x['DATASYNC_ID']), str) or '').split('||')
655 if len(sync_ids) >= 2 and sync_ids[1]:
656 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
657 # and just "user_syncid||" for primary channel. We only want the channel_syncid
658 return sync_ids[0]
659
660 @staticmethod
661 def _extract_visitor_data(*args):
662 """
663 Extracts visitorData from an API response or ytcfg
664 Appears to be used to track session state
665 """
666 return get_first(
667 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
668 expected_type=str)
669
670 @functools.cached_property
671 def is_authenticated(self):
672 return bool(self._generate_sapisidhash_header())
673
674 def extract_ytcfg(self, video_id, webpage):
675 if not webpage:
676 return {}
677 return self._parse_json(
678 self._search_regex(
679 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
680 default='{}'), video_id, fatal=False) or {}
681
682 def generate_api_headers(
683 self, *, ytcfg=None, account_syncid=None, session_index=None,
684 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
685
686 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
687 headers = {
688 'X-YouTube-Client-Name': str(
689 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
690 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
691 'Origin': origin,
692 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
693 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
694 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
695 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
696 }
697 if session_index is None:
698 session_index = self._extract_session_index(ytcfg)
699 if account_syncid or session_index is not None:
700 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
701
702 auth = self._generate_sapisidhash_header(origin)
703 if auth is not None:
704 headers['Authorization'] = auth
705 headers['X-Origin'] = origin
706 return filter_dict(headers)
707
708 def _download_ytcfg(self, client, video_id):
709 url = {
710 'web': 'https://www.youtube.com',
711 'web_music': 'https://music.youtube.com',
712 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
713 }.get(client)
714 if not url:
715 return {}
716 webpage = self._download_webpage(
717 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
718 return self.extract_ytcfg(video_id, webpage) or {}
719
720 @staticmethod
721 def _build_api_continuation_query(continuation, ctp=None):
722 query = {
723 'continuation': continuation
724 }
725 # TODO: Inconsistency with clickTrackingParams.
726 # Currently we have a fixed ctp contained within context (from ytcfg)
727 # and a ctp in root query for continuation.
728 if ctp:
729 query['clickTracking'] = {'clickTrackingParams': ctp}
730 return query
731
732 @classmethod
733 def _extract_next_continuation_data(cls, renderer):
734 next_continuation = try_get(
735 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
736 lambda x: x['continuation']['reloadContinuationData']), dict)
737 if not next_continuation:
738 return
739 continuation = next_continuation.get('continuation')
740 if not continuation:
741 return
742 ctp = next_continuation.get('clickTrackingParams')
743 return cls._build_api_continuation_query(continuation, ctp)
744
745 @classmethod
746 def _extract_continuation_ep_data(cls, continuation_ep: dict):
747 if isinstance(continuation_ep, dict):
748 continuation = try_get(
749 continuation_ep, lambda x: x['continuationCommand']['token'], str)
750 if not continuation:
751 return
752 ctp = continuation_ep.get('clickTrackingParams')
753 return cls._build_api_continuation_query(continuation, ctp)
754
755 @classmethod
756 def _extract_continuation(cls, renderer):
757 next_continuation = cls._extract_next_continuation_data(renderer)
758 if next_continuation:
759 return next_continuation
760
761 return traverse_obj(renderer, (
762 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
763 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
764 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
765
766 @classmethod
767 def _extract_alerts(cls, data):
768 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
769 if not isinstance(alert_dict, dict):
770 continue
771 for alert in alert_dict.values():
772 alert_type = alert.get('type')
773 if not alert_type:
774 continue
775 message = cls._get_text(alert, 'text')
776 if message:
777 yield alert_type, message
778
779 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
780 errors, warnings = [], []
781 for alert_type, alert_message in alerts:
782 if alert_type.lower() == 'error' and fatal:
783 errors.append([alert_type, alert_message])
784 elif alert_message not in self._IGNORED_WARNINGS:
785 warnings.append([alert_type, alert_message])
786
787 for alert_type, alert_message in (warnings + errors[:-1]):
788 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
789 if errors:
790 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
791
792 def _extract_and_report_alerts(self, data, *args, **kwargs):
793 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
794
795 def _extract_badges(self, badge_list: list):
796 """
797 Extract known BadgeType's from a list of badge renderers.
798 @returns [{'type': BadgeType}]
799 """
800 icon_type_map = {
801 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
802 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
803 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
804 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
805 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
806 'CHECK': BadgeType.VERIFIED,
807 }
808
809 badge_style_map = {
810 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
811 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
812 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
813 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
814 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
815 }
816
817 label_map = {
818 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
819 'private': BadgeType.AVAILABILITY_PRIVATE,
820 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
821 'live': BadgeType.LIVE_NOW,
822 'premium': BadgeType.AVAILABILITY_PREMIUM,
823 'verified': BadgeType.VERIFIED,
824 'official artist channel': BadgeType.VERIFIED,
825 }
826
827 badges = []
828 for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
829 badge_type = (
830 icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
831 or badge_style_map.get(traverse_obj(badge, 'style'))
832 )
833 if badge_type:
834 badges.append({'type': badge_type})
835 continue
836
837 # fallback, won't work in some languages
838 label = traverse_obj(
839 badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
840 for match, label_badge_type in label_map.items():
841 if match in label.lower():
842 badges.append({'type': label_badge_type})
843 break
844
845 return badges
846
847 @staticmethod
848 def _has_badge(badges, badge_type):
849 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
850
851 @staticmethod
852 def _get_text(data, *path_list, max_runs=None):
853 for path in path_list or [None]:
854 if path is None:
855 obj = [data]
856 else:
857 obj = traverse_obj(data, path, default=[])
858 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
859 obj = [obj]
860 for item in obj:
861 text = try_get(item, lambda x: x['simpleText'], str)
862 if text:
863 return text
864 runs = try_get(item, lambda x: x['runs'], list) or []
865 if not runs and isinstance(item, list):
866 runs = item
867
868 runs = runs[:min(len(runs), max_runs or len(runs))]
869 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
870 if text:
871 return text
872
873 def _get_count(self, data, *path_list):
874 count_text = self._get_text(data, *path_list) or ''
875 count = parse_count(count_text)
876 if count is None:
877 count = str_to_int(
878 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
879 return count
880
881 @staticmethod
882 def _extract_thumbnails(data, *path_list):
883 """
884 Extract thumbnails from thumbnails dict
885 @param path_list: path list to level that contains 'thumbnails' key
886 """
887 thumbnails = []
888 for path in path_list or [()]:
889 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
890 thumbnail_url = url_or_none(thumbnail.get('url'))
891 if not thumbnail_url:
892 continue
893 # Sometimes youtube gives a wrong thumbnail URL. See:
894 # https://github.com/yt-dlp/yt-dlp/issues/233
895 # https://github.com/ytdl-org/youtube-dl/issues/28023
896 if 'maxresdefault' in thumbnail_url:
897 thumbnail_url = thumbnail_url.split('?')[0]
898 thumbnails.append({
899 'url': thumbnail_url,
900 'height': int_or_none(thumbnail.get('height')),
901 'width': int_or_none(thumbnail.get('width')),
902 })
903 return thumbnails
904
905 @staticmethod
906 def extract_relative_time(relative_time_text):
907 """
908 Extracts a relative time from string and converts to dt object
909 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
910 """
911
912 # XXX: this could be moved to a general function in utils.py
913 # The relative time text strings are roughly the same as what
914 # Javascript's Intl.RelativeTimeFormat function generates.
915 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
916 mobj = re.search(
917 r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>sec(?:ond)?|s|min(?:ute)?|h(?:our|r)?|d(?:ay)?|w(?:eek|k)?|mo(?:nth)?|y(?:ear|r)?)s?\s*ago',
918 relative_time_text)
919 if mobj:
920 start = mobj.group('start')
921 if start:
922 return datetime_from_str(start)
923 try:
924 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
925 except ValueError:
926 return None
927
928 def _parse_time_text(self, text):
929 if not text:
930 return
931 dt = self.extract_relative_time(text)
932 timestamp = None
933 if isinstance(dt, datetime.datetime):
934 timestamp = calendar.timegm(dt.timetuple())
935
936 if timestamp is None:
937 timestamp = (
938 unified_timestamp(text) or unified_timestamp(
939 self._search_regex(
940 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
941 text.lower(), 'time text', default=None)))
942
943 if text and timestamp is None and self._preferred_lang in (None, 'en'):
944 self.report_warning(
945 f'Cannot parse localized time text "{text}"', only_once=True)
946 return timestamp
947
948 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
949 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
950 default_client='web'):
951 for retry in self.RetryManager():
952 try:
953 response = self._call_api(
954 ep=ep, fatal=True, headers=headers,
955 video_id=item_id, query=query, note=note,
956 context=self._extract_context(ytcfg, default_client),
957 api_key=self._extract_api_key(ytcfg, default_client),
958 api_hostname=api_hostname, default_client=default_client)
959 except ExtractorError as e:
960 if not isinstance(e.cause, network_exceptions):
961 return self._error_or_warning(e, fatal=fatal)
962 elif not isinstance(e.cause, urllib.error.HTTPError):
963 retry.error = e
964 continue
965
966 first_bytes = e.cause.read(512)
967 if not is_html(first_bytes):
968 yt_error = try_get(
969 self._parse_json(
970 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
971 lambda x: x['error']['message'], str)
972 if yt_error:
973 self._report_alerts([('ERROR', yt_error)], fatal=False)
974 # Downloading page may result in intermittent 5xx HTTP error
975 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
976 # We also want to catch all other network exceptions since errors in later pages can be troublesome
977 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
978 if e.cause.code not in (403, 429):
979 retry.error = e
980 continue
981 return self._error_or_warning(e, fatal=fatal)
982
983 try:
984 self._extract_and_report_alerts(response, only_once=True)
985 except ExtractorError as e:
986 # YouTube servers may return errors we want to retry on in a 200 OK response
987 # See: https://github.com/yt-dlp/yt-dlp/issues/839
988 if 'unknown error' in e.msg.lower():
989 retry.error = e
990 continue
991 return self._error_or_warning(e, fatal=fatal)
992 # Youtube sometimes sends incomplete data
993 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
994 if not traverse_obj(response, *variadic(check_get_keys)):
995 retry.error = ExtractorError('Incomplete data received', expected=True)
996 continue
997
998 return response
999
1000 @staticmethod
1001 def is_music_url(url):
1002 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1003
1004 def _extract_video(self, renderer):
1005 video_id = renderer.get('videoId')
1006
1007 reel_header_renderer = traverse_obj(renderer, (
1008 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1009 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1010
1011 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1012 description = self._get_text(renderer, 'descriptionSnippet')
1013
1014 duration = int_or_none(renderer.get('lengthSeconds'))
1015 if duration is None:
1016 duration = parse_duration(self._get_text(
1017 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1018 if duration is None:
1019 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1020 duration = parse_duration(self._search_regex(
1021 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1022 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1023 video_id, default=None, group='duration'))
1024
1025 channel_id = traverse_obj(
1026 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1027 expected_type=str, get_all=False)
1028 if not channel_id:
1029 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1030
1031 channel_id = self.ucid_or_none(channel_id)
1032
1033 overlay_style = traverse_obj(
1034 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1035 get_all=False, expected_type=str)
1036 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1037 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1038 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1039 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1040 expected_type=str)) or ''
1041 url = f'https://www.youtube.com/watch?v={video_id}'
1042 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1043 url = f'https://www.youtube.com/shorts/{video_id}'
1044
1045 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1046 or self._get_text(reel_header_renderer, 'timestampText') or '')
1047 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1048
1049 live_status = (
1050 'is_upcoming' if scheduled_timestamp is not None
1051 else 'was_live' if 'streamed' in time_text.lower()
1052 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1053 else None)
1054
1055 # videoInfo is a string like '50K views • 10 years ago'.
1056 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1057 view_count = (0 if 'no views' in view_count_text.lower()
1058 else self._get_count({'simpleText': view_count_text}))
1059 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1060
1061 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1062 or self._get_text(reel_header_renderer, 'channelTitleText'))
1063
1064 channel_handle = traverse_obj(renderer, (
1065 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1066 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1067 expected_type=self.handle_from_url, get_all=False)
1068 return {
1069 '_type': 'url',
1070 'ie_key': YoutubeIE.ie_key(),
1071 'id': video_id,
1072 'url': url,
1073 'title': title,
1074 'description': description,
1075 'duration': duration,
1076 'channel_id': channel_id,
1077 'channel': channel,
1078 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1079 'uploader': channel,
1080 'uploader_id': channel_handle,
1081 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1082 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1083 'timestamp': (self._parse_time_text(time_text)
1084 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1085 else None),
1086 'release_timestamp': scheduled_timestamp,
1087 'availability':
1088 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1089 else self._availability(
1090 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1091 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1092 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1093 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1094 view_count_field: view_count,
1095 'live_status': live_status,
1096 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
1097 }
1098
1099
1100 class YoutubeIE(YoutubeBaseInfoExtractor):
1101 IE_DESC = 'YouTube'
1102 _VALID_URL = r"""(?x)^
1103 (
1104 (?:https?://|//) # http(s):// or protocol-independent URL
1105 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1106 (?:www\.)?deturl\.com/www\.youtube\.com|
1107 (?:www\.)?pwnyoutube\.com|
1108 (?:www\.)?hooktube\.com|
1109 (?:www\.)?yourepeat\.com|
1110 tube\.majestyc\.net|
1111 %(invidious)s|
1112 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1113 (?:.*?\#/)? # handle anchor (#/) redirect urls
1114 (?: # the various things that can precede the ID:
1115 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1116 |(?: # or the v= param in all its forms
1117 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1118 (?:\?|\#!?) # the params delimiter ? or # or #!
1119 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1120 v=
1121 )
1122 ))
1123 |(?:
1124 youtu\.be| # just youtu.be/xxxx
1125 vid\.plus| # or vid.plus/xxxx
1126 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1127 %(invidious)s
1128 )/
1129 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1130 )
1131 )? # all until now is optional -> you can pass the naked ID
1132 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1133 (?(1).+)? # if we found the ID, everything can follow
1134 (?:\#|$)""" % {
1135 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1136 }
1137 _EMBED_REGEX = [
1138 r'''(?x)
1139 (?:
1140 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1141 data-video-url=|
1142 <embed[^>]+?src=|
1143 embedSWF\(?:\s*|
1144 <object[^>]+data=|
1145 new\s+SWFObject\(
1146 )
1147 (["\'])
1148 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1149 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1150 \1''',
1151 # https://wordpress.org/plugins/lazy-load-for-videos/
1152 r'''(?xs)
1153 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1154 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1155 ]
1156 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1157
1158 _PLAYER_INFO_RE = (
1159 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1160 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
1161 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1162 )
1163 _formats = {
1164 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1165 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1166 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1167 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1168 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1169 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1170 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1171 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1172 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1173 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1174 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1175 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1176 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1177 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1178 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1179 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1180 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1181 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1182
1183
1184 # 3D videos
1185 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1186 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1187 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1188 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1189 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1190 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1191 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1192
1193 # Apple HTTP Live Streaming
1194 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1195 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1196 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1197 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1198 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1199 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1200 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1201 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1202
1203 # DASH mp4 video
1204 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1205 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1206 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1207 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1208 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1209 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1210 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1211 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1212 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1213 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1214 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1215 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1216
1217 # Dash mp4 audio
1218 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1219 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1220 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1221 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1222 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1223 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1224 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1225
1226 # Dash webm
1227 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1228 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1229 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1230 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1231 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1232 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1233 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1234 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1235 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1236 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1237 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1238 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1239 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1240 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1241 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1242 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1243 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1244 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1245 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1246 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1247 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1248 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1249
1250 # Dash webm audio
1251 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1252 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1253
1254 # Dash webm audio with opus inside
1255 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1256 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1257 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1258
1259 # RTMP (unnamed)
1260 '_rtmp': {'protocol': 'rtmp'},
1261
1262 # av01 video only formats sometimes served with "unknown" codecs
1263 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1264 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1265 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1266 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1267 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1268 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1269 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1270 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1271 }
1272 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1273
1274 _GEO_BYPASS = False
1275
1276 IE_NAME = 'youtube'
1277 _TESTS = [
1278 {
1279 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1280 'info_dict': {
1281 'id': 'BaW_jenozKc',
1282 'ext': 'mp4',
1283 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1284 'channel': 'Philipp Hagemeister',
1285 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1286 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1287 'upload_date': '20121002',
1288 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1289 'categories': ['Science & Technology'],
1290 'tags': ['youtube-dl'],
1291 'duration': 10,
1292 'view_count': int,
1293 'like_count': int,
1294 'availability': 'public',
1295 'playable_in_embed': True,
1296 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1297 'live_status': 'not_live',
1298 'age_limit': 0,
1299 'start_time': 1,
1300 'end_time': 9,
1301 'comment_count': int,
1302 'channel_follower_count': int,
1303 'uploader': 'Philipp Hagemeister',
1304 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1305 'uploader_id': '@PhilippHagemeister',
1306 'heatmap': 'count:100',
1307 }
1308 },
1309 {
1310 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1311 'note': 'Embed-only video (#1746)',
1312 'info_dict': {
1313 'id': 'yZIXLfi8CZQ',
1314 'ext': 'mp4',
1315 'upload_date': '20120608',
1316 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1317 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1318 'age_limit': 18,
1319 },
1320 'skip': 'Private video',
1321 },
1322 {
1323 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1324 'note': 'Use the first video ID in the URL',
1325 'info_dict': {
1326 'id': 'BaW_jenozKc',
1327 'ext': 'mp4',
1328 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1329 'channel': 'Philipp Hagemeister',
1330 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1331 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1332 'upload_date': '20121002',
1333 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1334 'categories': ['Science & Technology'],
1335 'tags': ['youtube-dl'],
1336 'duration': 10,
1337 'view_count': int,
1338 'like_count': int,
1339 'availability': 'public',
1340 'playable_in_embed': True,
1341 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1342 'live_status': 'not_live',
1343 'age_limit': 0,
1344 'comment_count': int,
1345 'channel_follower_count': int,
1346 'uploader': 'Philipp Hagemeister',
1347 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1348 'uploader_id': '@PhilippHagemeister',
1349 'heatmap': 'count:100',
1350 },
1351 'params': {
1352 'skip_download': True,
1353 },
1354 },
1355 {
1356 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1357 'note': '256k DASH audio (format 141) via DASH manifest',
1358 'info_dict': {
1359 'id': 'a9LDPn-MO4I',
1360 'ext': 'm4a',
1361 'upload_date': '20121002',
1362 'description': '',
1363 'title': 'UHDTV TEST 8K VIDEO.mp4'
1364 },
1365 'params': {
1366 'youtube_include_dash_manifest': True,
1367 'format': '141',
1368 },
1369 'skip': 'format 141 not served anymore',
1370 },
1371 # DASH manifest with encrypted signature
1372 {
1373 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1374 'info_dict': {
1375 'id': 'IB3lcPjvWLA',
1376 'ext': 'm4a',
1377 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1378 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1379 'duration': 244,
1380 'upload_date': '20131011',
1381 'abr': 129.495,
1382 'like_count': int,
1383 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1384 'playable_in_embed': True,
1385 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1386 'view_count': int,
1387 'track': 'The Spark',
1388 'live_status': 'not_live',
1389 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1390 'channel': 'Afrojack',
1391 'tags': 'count:19',
1392 'availability': 'public',
1393 'categories': ['Music'],
1394 'age_limit': 0,
1395 'alt_title': 'The Spark',
1396 'channel_follower_count': int,
1397 'uploader': 'Afrojack',
1398 'uploader_url': 'https://www.youtube.com/@Afrojack',
1399 'uploader_id': '@Afrojack',
1400 },
1401 'params': {
1402 'youtube_include_dash_manifest': True,
1403 'format': '141/bestaudio[ext=m4a]',
1404 },
1405 },
1406 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1407 {
1408 'note': 'Embed allowed age-gate video',
1409 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1410 'info_dict': {
1411 'id': 'HtVdAasjOgU',
1412 'ext': 'mp4',
1413 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1414 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1415 'duration': 142,
1416 'upload_date': '20140605',
1417 'age_limit': 18,
1418 'categories': ['Gaming'],
1419 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1420 'availability': 'needs_auth',
1421 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1422 'like_count': int,
1423 'channel': 'The Witcher',
1424 'live_status': 'not_live',
1425 'tags': 'count:17',
1426 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1427 'playable_in_embed': True,
1428 'view_count': int,
1429 'channel_follower_count': int,
1430 'uploader': 'The Witcher',
1431 'uploader_url': 'https://www.youtube.com/@thewitcher',
1432 'uploader_id': '@thewitcher',
1433 'comment_count': int,
1434 'channel_is_verified': True,
1435 'heatmap': 'count:100',
1436 },
1437 },
1438 {
1439 'note': 'Age-gate video with embed allowed in public site',
1440 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1441 'info_dict': {
1442 'id': 'HsUATh_Nc2U',
1443 'ext': 'mp4',
1444 'title': 'Godzilla 2 (Official Video)',
1445 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1446 'upload_date': '20200408',
1447 'age_limit': 18,
1448 'availability': 'needs_auth',
1449 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1450 'channel': 'FlyingKitty',
1451 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1452 'view_count': int,
1453 'categories': ['Entertainment'],
1454 'live_status': 'not_live',
1455 'tags': ['Flyingkitty', 'godzilla 2'],
1456 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1457 'like_count': int,
1458 'duration': 177,
1459 'playable_in_embed': True,
1460 'channel_follower_count': int,
1461 'uploader': 'FlyingKitty',
1462 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1463 'uploader_id': '@FlyingKitty900',
1464 'comment_count': int,
1465 'channel_is_verified': True,
1466 },
1467 },
1468 {
1469 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1470 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1471 'info_dict': {
1472 'id': 'Tq92D6wQ1mg',
1473 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1474 'ext': 'mp4',
1475 'upload_date': '20191228',
1476 'description': 'md5:17eccca93a786d51bc67646756894066',
1477 'age_limit': 18,
1478 'like_count': int,
1479 'availability': 'needs_auth',
1480 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1481 'view_count': int,
1482 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1483 'channel': 'Projekt Melody',
1484 'live_status': 'not_live',
1485 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1486 'playable_in_embed': True,
1487 'categories': ['Entertainment'],
1488 'duration': 106,
1489 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1490 'comment_count': int,
1491 'channel_follower_count': int,
1492 'uploader': 'Projekt Melody',
1493 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1494 'uploader_id': '@ProjektMelody',
1495 },
1496 },
1497 {
1498 'note': 'Non-Agegated non-embeddable video',
1499 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1500 'info_dict': {
1501 'id': 'MeJVWBSsPAY',
1502 'ext': 'mp4',
1503 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1504 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1505 'upload_date': '20130730',
1506 'track': 'Such mich find mich',
1507 'age_limit': 0,
1508 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1509 'like_count': int,
1510 'playable_in_embed': False,
1511 'creator': 'OOMPH!',
1512 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1513 'view_count': int,
1514 'alt_title': 'Such mich find mich',
1515 'duration': 210,
1516 'channel': 'Herr Lurik',
1517 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1518 'categories': ['Music'],
1519 'availability': 'public',
1520 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1521 'live_status': 'not_live',
1522 'artist': 'OOMPH!',
1523 'channel_follower_count': int,
1524 'uploader': 'Herr Lurik',
1525 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1526 'uploader_id': '@HerrLurik',
1527 },
1528 },
1529 {
1530 'note': 'Non-bypassable age-gated video',
1531 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1532 'only_matching': True,
1533 },
1534 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1535 # YouTube Red ad is not captured for creator
1536 {
1537 'url': '__2ABJjxzNo',
1538 'info_dict': {
1539 'id': '__2ABJjxzNo',
1540 'ext': 'mp4',
1541 'duration': 266,
1542 'upload_date': '20100430',
1543 'creator': 'deadmau5',
1544 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1545 'title': 'Deadmau5 - Some Chords (HD)',
1546 'alt_title': 'Some Chords',
1547 'availability': 'public',
1548 'tags': 'count:14',
1549 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1550 'view_count': int,
1551 'live_status': 'not_live',
1552 'channel': 'deadmau5',
1553 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1554 'like_count': int,
1555 'track': 'Some Chords',
1556 'artist': 'deadmau5',
1557 'playable_in_embed': True,
1558 'age_limit': 0,
1559 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1560 'categories': ['Music'],
1561 'album': 'Some Chords',
1562 'channel_follower_count': int,
1563 'uploader': 'deadmau5',
1564 'uploader_url': 'https://www.youtube.com/@deadmau5',
1565 'uploader_id': '@deadmau5',
1566 },
1567 'expected_warnings': [
1568 'DASH manifest missing',
1569 ]
1570 },
1571 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1572 {
1573 'url': 'lqQg6PlCWgI',
1574 'info_dict': {
1575 'id': 'lqQg6PlCWgI',
1576 'ext': 'mp4',
1577 'duration': 6085,
1578 'upload_date': '20150827',
1579 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1580 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1581 'like_count': int,
1582 'release_timestamp': 1343767800,
1583 'playable_in_embed': True,
1584 'categories': ['Sports'],
1585 'release_date': '20120731',
1586 'channel': 'Olympics',
1587 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1588 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1589 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1590 'age_limit': 0,
1591 'availability': 'public',
1592 'live_status': 'was_live',
1593 'view_count': int,
1594 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1595 'channel_follower_count': int,
1596 'uploader': 'Olympics',
1597 'uploader_url': 'https://www.youtube.com/@Olympics',
1598 'uploader_id': '@Olympics',
1599 'channel_is_verified': True,
1600 },
1601 'params': {
1602 'skip_download': 'requires avconv',
1603 }
1604 },
1605 # Non-square pixels
1606 {
1607 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1608 'info_dict': {
1609 'id': '_b-2C3KPAM0',
1610 'ext': 'mp4',
1611 'stretched_ratio': 16 / 9.,
1612 'duration': 85,
1613 'upload_date': '20110310',
1614 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
1615 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1616 'playable_in_embed': True,
1617 'channel': '孫ᄋᄅ',
1618 'age_limit': 0,
1619 'tags': 'count:11',
1620 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1621 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1622 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1623 'view_count': int,
1624 'categories': ['People & Blogs'],
1625 'like_count': int,
1626 'live_status': 'not_live',
1627 'availability': 'unlisted',
1628 'comment_count': int,
1629 'channel_follower_count': int,
1630 'uploader': '孫ᄋᄅ',
1631 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1632 'uploader_id': '@AllenMeow',
1633 },
1634 },
1635 # url_encoded_fmt_stream_map is empty string
1636 {
1637 'url': 'qEJwOuvDf7I',
1638 'info_dict': {
1639 'id': 'qEJwOuvDf7I',
1640 'ext': 'webm',
1641 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1642 'description': '',
1643 'upload_date': '20150404',
1644 },
1645 'params': {
1646 'skip_download': 'requires avconv',
1647 },
1648 'skip': 'This live event has ended.',
1649 },
1650 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1651 {
1652 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1653 'info_dict': {
1654 'id': 'FIl7x6_3R5Y',
1655 'ext': 'webm',
1656 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1657 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1658 'duration': 220,
1659 'upload_date': '20150625',
1660 'formats': 'mincount:31',
1661 },
1662 'skip': 'not actual anymore',
1663 },
1664 # DASH manifest with segment_list
1665 {
1666 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1667 'md5': '8ce563a1d667b599d21064e982ab9e31',
1668 'info_dict': {
1669 'id': 'CsmdDsKjzN8',
1670 'ext': 'mp4',
1671 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1672 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1673 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1674 },
1675 'params': {
1676 'youtube_include_dash_manifest': True,
1677 'format': '135', # bestvideo
1678 },
1679 'skip': 'This live event has ended.',
1680 },
1681 {
1682 # Multifeed videos (multiple cameras), URL can be of any Camera
1683 # TODO: fix multifeed titles
1684 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1685 'info_dict': {
1686 'id': 'zaPI8MvL8pg',
1687 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1688 'description': 'md5:563ccbc698b39298481ca3c571169519',
1689 },
1690 'playlist': [{
1691 'info_dict': {
1692 'id': 'j5yGuxZ8lLU',
1693 'ext': 'mp4',
1694 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1695 'description': 'md5:563ccbc698b39298481ca3c571169519',
1696 'duration': 10120,
1697 'channel_follower_count': int,
1698 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1699 'availability': 'public',
1700 'playable_in_embed': True,
1701 'upload_date': '20131105',
1702 'categories': ['Gaming'],
1703 'live_status': 'was_live',
1704 'tags': 'count:24',
1705 'release_timestamp': 1383701910,
1706 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1707 'comment_count': int,
1708 'age_limit': 0,
1709 'like_count': int,
1710 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1711 'channel': 'WiiLikeToPlay',
1712 'view_count': int,
1713 'release_date': '20131106',
1714 'uploader': 'WiiLikeToPlay',
1715 'uploader_id': '@WLTP',
1716 'uploader_url': 'https://www.youtube.com/@WLTP',
1717 },
1718 }, {
1719 'info_dict': {
1720 'id': 'zaPI8MvL8pg',
1721 'ext': 'mp4',
1722 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1723 'availability': 'public',
1724 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1725 'channel': 'WiiLikeToPlay',
1726 'channel_follower_count': int,
1727 'description': 'md5:563ccbc698b39298481ca3c571169519',
1728 'duration': 10108,
1729 'age_limit': 0,
1730 'like_count': int,
1731 'tags': 'count:24',
1732 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1733 'release_timestamp': 1383701915,
1734 'comment_count': int,
1735 'upload_date': '20131105',
1736 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1737 'release_date': '20131106',
1738 'playable_in_embed': True,
1739 'live_status': 'was_live',
1740 'categories': ['Gaming'],
1741 'view_count': int,
1742 'uploader': 'WiiLikeToPlay',
1743 'uploader_id': '@WLTP',
1744 'uploader_url': 'https://www.youtube.com/@WLTP',
1745 },
1746 }, {
1747 'info_dict': {
1748 'id': 'R7r3vfO7Hao',
1749 'ext': 'mp4',
1750 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1751 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1752 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1753 'like_count': int,
1754 'availability': 'public',
1755 'playable_in_embed': True,
1756 'upload_date': '20131105',
1757 'description': 'md5:563ccbc698b39298481ca3c571169519',
1758 'channel_follower_count': int,
1759 'tags': 'count:24',
1760 'release_date': '20131106',
1761 'comment_count': int,
1762 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1763 'channel': 'WiiLikeToPlay',
1764 'categories': ['Gaming'],
1765 'release_timestamp': 1383701914,
1766 'live_status': 'was_live',
1767 'age_limit': 0,
1768 'duration': 10128,
1769 'view_count': int,
1770 'uploader': 'WiiLikeToPlay',
1771 'uploader_id': '@WLTP',
1772 'uploader_url': 'https://www.youtube.com/@WLTP',
1773 },
1774 }],
1775 'params': {'skip_download': True},
1776 },
1777 {
1778 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1779 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1780 'info_dict': {
1781 'id': 'gVfLd0zydlo',
1782 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1783 },
1784 'playlist_count': 2,
1785 'skip': 'Not multifeed anymore',
1786 },
1787 {
1788 'url': 'https://vid.plus/FlRa-iH7PGw',
1789 'only_matching': True,
1790 },
1791 {
1792 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1793 'only_matching': True,
1794 },
1795 {
1796 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1797 # Also tests cut-off URL expansion in video description (see
1798 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1799 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1800 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1801 'info_dict': {
1802 'id': 'lsguqyKfVQg',
1803 'ext': 'mp4',
1804 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1805 'alt_title': 'Dark Walk',
1806 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1807 'duration': 133,
1808 'upload_date': '20151119',
1809 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1810 'track': 'Dark Walk',
1811 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1812 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1813 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1814 'categories': ['Film & Animation'],
1815 'view_count': int,
1816 'live_status': 'not_live',
1817 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1818 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1819 'tags': 'count:13',
1820 'availability': 'public',
1821 'channel': 'IronSoulElf',
1822 'playable_in_embed': True,
1823 'like_count': int,
1824 'age_limit': 0,
1825 'channel_follower_count': int
1826 },
1827 'params': {
1828 'skip_download': True,
1829 },
1830 },
1831 {
1832 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1833 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1834 'only_matching': True,
1835 },
1836 {
1837 # Video with yt:stretch=17:0
1838 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1839 'info_dict': {
1840 'id': 'Q39EVAstoRM',
1841 'ext': 'mp4',
1842 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1843 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1844 'upload_date': '20151107',
1845 },
1846 'params': {
1847 'skip_download': True,
1848 },
1849 'skip': 'This video does not exist.',
1850 },
1851 {
1852 # Video with incomplete 'yt:stretch=16:'
1853 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1854 'only_matching': True,
1855 },
1856 {
1857 # Video licensed under Creative Commons
1858 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1859 'info_dict': {
1860 'id': 'M4gD1WSo5mA',
1861 'ext': 'mp4',
1862 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1863 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1864 'duration': 721,
1865 'upload_date': '20150128',
1866 'license': 'Creative Commons Attribution license (reuse allowed)',
1867 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1868 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1869 'like_count': int,
1870 'age_limit': 0,
1871 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1872 'channel': 'The Berkman Klein Center for Internet & Society',
1873 'availability': 'public',
1874 'view_count': int,
1875 'categories': ['Education'],
1876 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1877 'live_status': 'not_live',
1878 'playable_in_embed': True,
1879 'channel_follower_count': int,
1880 'chapters': list,
1881 'uploader': 'The Berkman Klein Center for Internet & Society',
1882 'uploader_id': '@BKCHarvard',
1883 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1884 },
1885 'params': {
1886 'skip_download': True,
1887 },
1888 },
1889 {
1890 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1891 'info_dict': {
1892 'id': 'eQcmzGIKrzg',
1893 'ext': 'mp4',
1894 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1895 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1896 'duration': 4060,
1897 'upload_date': '20151120',
1898 'license': 'Creative Commons Attribution license (reuse allowed)',
1899 'playable_in_embed': True,
1900 'tags': 'count:12',
1901 'like_count': int,
1902 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1903 'age_limit': 0,
1904 'availability': 'public',
1905 'categories': ['News & Politics'],
1906 'channel': 'Bernie Sanders',
1907 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1908 'view_count': int,
1909 'live_status': 'not_live',
1910 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1911 'comment_count': int,
1912 'channel_follower_count': int,
1913 'chapters': list,
1914 'uploader': 'Bernie Sanders',
1915 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1916 'uploader_id': '@BernieSanders',
1917 'channel_is_verified': True,
1918 'heatmap': 'count:100',
1919 },
1920 'params': {
1921 'skip_download': True,
1922 },
1923 },
1924 {
1925 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1926 'only_matching': True,
1927 },
1928 {
1929 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1930 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1931 'only_matching': True,
1932 },
1933 {
1934 # Rental video preview
1935 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1936 'info_dict': {
1937 'id': 'uGpuVWrhIzE',
1938 'ext': 'mp4',
1939 'title': 'Piku - Trailer',
1940 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1941 'upload_date': '20150811',
1942 'license': 'Standard YouTube License',
1943 },
1944 'params': {
1945 'skip_download': True,
1946 },
1947 'skip': 'This video is not available.',
1948 },
1949 {
1950 # YouTube Red video with episode data
1951 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1952 'info_dict': {
1953 'id': 'iqKdEhx-dD4',
1954 'ext': 'mp4',
1955 'title': 'Isolation - Mind Field (Ep 1)',
1956 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1957 'duration': 2085,
1958 'upload_date': '20170118',
1959 'series': 'Mind Field',
1960 'season_number': 1,
1961 'episode_number': 1,
1962 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1963 'tags': 'count:12',
1964 'view_count': int,
1965 'availability': 'public',
1966 'age_limit': 0,
1967 'channel': 'Vsauce',
1968 'episode': 'Episode 1',
1969 'categories': ['Entertainment'],
1970 'season': 'Season 1',
1971 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1972 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1973 'like_count': int,
1974 'playable_in_embed': True,
1975 'live_status': 'not_live',
1976 'channel_follower_count': int,
1977 'uploader': 'Vsauce',
1978 'uploader_url': 'https://www.youtube.com/@Vsauce',
1979 'uploader_id': '@Vsauce',
1980 'comment_count': int,
1981 'channel_is_verified': True,
1982 },
1983 'params': {
1984 'skip_download': True,
1985 },
1986 'expected_warnings': [
1987 'Skipping DASH manifest',
1988 ],
1989 },
1990 {
1991 # The following content has been identified by the YouTube community
1992 # as inappropriate or offensive to some audiences.
1993 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1994 'info_dict': {
1995 'id': '6SJNVb0GnPI',
1996 'ext': 'mp4',
1997 'title': 'Race Differences in Intelligence',
1998 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1999 'duration': 965,
2000 'upload_date': '20140124',
2001 },
2002 'params': {
2003 'skip_download': True,
2004 },
2005 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2006 },
2007 {
2008 # itag 212
2009 'url': '1t24XAntNCY',
2010 'only_matching': True,
2011 },
2012 {
2013 # geo restricted to JP
2014 'url': 'sJL6WA-aGkQ',
2015 'only_matching': True,
2016 },
2017 {
2018 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2019 'only_matching': True,
2020 },
2021 {
2022 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2023 'only_matching': True,
2024 },
2025 {
2026 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2027 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2028 'only_matching': True,
2029 },
2030 {
2031 # DRM protected
2032 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2033 'only_matching': True,
2034 },
2035 {
2036 # Video with unsupported adaptive stream type formats
2037 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2038 'info_dict': {
2039 'id': 'Z4Vy8R84T1U',
2040 'ext': 'mp4',
2041 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2042 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2043 'duration': 433,
2044 'upload_date': '20130923',
2045 'formats': 'maxcount:10',
2046 },
2047 'params': {
2048 'skip_download': True,
2049 'youtube_include_dash_manifest': False,
2050 },
2051 'skip': 'not actual anymore',
2052 },
2053 {
2054 # Youtube Music Auto-generated description
2055 # TODO: fix metadata extraction
2056 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2057 'info_dict': {
2058 'id': 'MgNrAu2pzNs',
2059 'ext': 'mp4',
2060 'title': 'Voyeur Girl',
2061 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2062 'upload_date': '20190312',
2063 'artist': 'Stephen',
2064 'track': 'Voyeur Girl',
2065 'album': 'it\'s too much love to know my dear',
2066 'release_date': '20190313',
2067 'release_year': 2019,
2068 'alt_title': 'Voyeur Girl',
2069 'view_count': int,
2070 'playable_in_embed': True,
2071 'like_count': int,
2072 'categories': ['Music'],
2073 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2074 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2075 'uploader': 'Stephen',
2076 'availability': 'public',
2077 'creator': 'Stephen',
2078 'duration': 169,
2079 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2080 'age_limit': 0,
2081 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2082 'tags': 'count:11',
2083 'live_status': 'not_live',
2084 'channel_follower_count': int
2085 },
2086 'params': {
2087 'skip_download': True,
2088 },
2089 },
2090 {
2091 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2092 'only_matching': True,
2093 },
2094 {
2095 # invalid -> valid video id redirection
2096 'url': 'DJztXj2GPfl',
2097 'info_dict': {
2098 'id': 'DJztXj2GPfk',
2099 'ext': 'mp4',
2100 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2101 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2102 'upload_date': '20090125',
2103 'artist': 'Panjabi MC',
2104 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2105 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2106 },
2107 'params': {
2108 'skip_download': True,
2109 },
2110 'skip': 'Video unavailable',
2111 },
2112 {
2113 # empty description results in an empty string
2114 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2115 'info_dict': {
2116 'id': 'x41yOUIvK2k',
2117 'ext': 'mp4',
2118 'title': 'IMG 3456',
2119 'description': '',
2120 'upload_date': '20170613',
2121 'view_count': int,
2122 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2123 'like_count': int,
2124 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2125 'tags': [],
2126 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2127 'availability': 'public',
2128 'age_limit': 0,
2129 'categories': ['Pets & Animals'],
2130 'duration': 7,
2131 'playable_in_embed': True,
2132 'live_status': 'not_live',
2133 'channel': 'l\'Or Vert asbl',
2134 'channel_follower_count': int,
2135 'uploader': 'l\'Or Vert asbl',
2136 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2137 'uploader_id': '@ElevageOrVert',
2138 },
2139 'params': {
2140 'skip_download': True,
2141 },
2142 },
2143 {
2144 # with '};' inside yt initial data (see [1])
2145 # see [2] for an example with '};' inside ytInitialPlayerResponse
2146 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2147 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2148 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2149 'info_dict': {
2150 'id': 'CHqg6qOn4no',
2151 'ext': 'mp4',
2152 'title': 'Part 77 Sort a list of simple types in c#',
2153 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2154 'upload_date': '20130831',
2155 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2156 'like_count': int,
2157 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2158 'live_status': 'not_live',
2159 'categories': ['Education'],
2160 'availability': 'public',
2161 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2162 'tags': 'count:12',
2163 'playable_in_embed': True,
2164 'age_limit': 0,
2165 'view_count': int,
2166 'duration': 522,
2167 'channel': 'kudvenkat',
2168 'comment_count': int,
2169 'channel_follower_count': int,
2170 'chapters': list,
2171 'uploader': 'kudvenkat',
2172 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2173 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2174 'channel_is_verified': True,
2175 'heatmap': 'count:100',
2176 },
2177 'params': {
2178 'skip_download': True,
2179 },
2180 },
2181 {
2182 # another example of '};' in ytInitialData
2183 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2184 'only_matching': True,
2185 },
2186 {
2187 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2188 'only_matching': True,
2189 },
2190 {
2191 # https://github.com/ytdl-org/youtube-dl/pull/28094
2192 'url': 'OtqTfy26tG0',
2193 'info_dict': {
2194 'id': 'OtqTfy26tG0',
2195 'ext': 'mp4',
2196 'title': 'Burn Out',
2197 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2198 'upload_date': '20141120',
2199 'artist': 'The Cinematic Orchestra',
2200 'track': 'Burn Out',
2201 'album': 'Every Day',
2202 'like_count': int,
2203 'live_status': 'not_live',
2204 'alt_title': 'Burn Out',
2205 'duration': 614,
2206 'age_limit': 0,
2207 'view_count': int,
2208 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2209 'creator': 'The Cinematic Orchestra',
2210 'channel': 'The Cinematic Orchestra',
2211 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2212 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2213 'availability': 'public',
2214 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2215 'categories': ['Music'],
2216 'playable_in_embed': True,
2217 'channel_follower_count': int,
2218 'uploader': 'The Cinematic Orchestra',
2219 'comment_count': int,
2220 },
2221 'params': {
2222 'skip_download': True,
2223 },
2224 },
2225 {
2226 # controversial video, only works with bpctr when authenticated with cookies
2227 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2228 'only_matching': True,
2229 },
2230 {
2231 # controversial video, requires bpctr/contentCheckOk
2232 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2233 'info_dict': {
2234 'id': 'SZJvDhaSDnc',
2235 'ext': 'mp4',
2236 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2237 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2238 'upload_date': '20140716',
2239 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2240 'duration': 170,
2241 'categories': ['News & Politics'],
2242 'view_count': int,
2243 'channel': 'CBS Mornings',
2244 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2245 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2246 'age_limit': 18,
2247 'availability': 'needs_auth',
2248 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2249 'like_count': int,
2250 'live_status': 'not_live',
2251 'playable_in_embed': True,
2252 'channel_follower_count': int,
2253 'uploader': 'CBS Mornings',
2254 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2255 'uploader_id': '@CBSMornings',
2256 'comment_count': int,
2257 'channel_is_verified': True,
2258 }
2259 },
2260 {
2261 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2262 'url': 'cBvYw8_A0vQ',
2263 'info_dict': {
2264 'id': 'cBvYw8_A0vQ',
2265 'ext': 'mp4',
2266 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2267 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2268 'upload_date': '20201120',
2269 'duration': 1456,
2270 'categories': ['Travel & Events'],
2271 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2272 'view_count': int,
2273 'channel': 'Walk around Japan',
2274 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2275 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2276 'age_limit': 0,
2277 'availability': 'public',
2278 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2279 'live_status': 'not_live',
2280 'playable_in_embed': True,
2281 'channel_follower_count': int,
2282 'uploader': 'Walk around Japan',
2283 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2284 'uploader_id': '@walkaroundjapan7124',
2285 },
2286 'params': {
2287 'skip_download': True,
2288 },
2289 }, {
2290 # Has multiple audio streams
2291 'url': 'WaOKSUlf4TM',
2292 'only_matching': True
2293 }, {
2294 # Requires Premium: has format 141 when requested using YTM url
2295 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2296 'only_matching': True
2297 }, {
2298 # multiple subtitles with same lang_code
2299 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2300 'only_matching': True,
2301 }, {
2302 # Force use android client fallback
2303 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2304 'info_dict': {
2305 'id': 'YOelRv7fMxY',
2306 'title': 'DIGGING A SECRET TUNNEL Part 1',
2307 'ext': '3gp',
2308 'upload_date': '20210624',
2309 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2310 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2311 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2312 'duration': 596,
2313 'categories': ['Entertainment'],
2314 'view_count': int,
2315 'channel': 'colinfurze',
2316 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2317 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2318 'age_limit': 0,
2319 'availability': 'public',
2320 'like_count': int,
2321 'live_status': 'not_live',
2322 'playable_in_embed': True,
2323 'channel_follower_count': int,
2324 'chapters': list,
2325 'uploader': 'colinfurze',
2326 'uploader_url': 'https://www.youtube.com/@colinfurze',
2327 'uploader_id': '@colinfurze',
2328 'comment_count': int,
2329 'channel_is_verified': True,
2330 'heatmap': 'count:100',
2331 },
2332 'params': {
2333 'format': '17', # 3gp format available on android
2334 'extractor_args': {'youtube': {'player_client': ['android']}},
2335 },
2336 },
2337 {
2338 # Skip download of additional client configs (remix client config in this case)
2339 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2340 'only_matching': True,
2341 'params': {
2342 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2343 },
2344 }, {
2345 # shorts
2346 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2347 'only_matching': True,
2348 }, {
2349 'note': 'Storyboards',
2350 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2351 'info_dict': {
2352 'id': '5KLPxDtMqe8',
2353 'ext': 'mhtml',
2354 'format_id': 'sb0',
2355 'title': 'Your Brain is Plastic',
2356 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2357 'upload_date': '20140324',
2358 'like_count': int,
2359 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2360 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2361 'view_count': int,
2362 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2363 'playable_in_embed': True,
2364 'tags': 'count:12',
2365 'availability': 'public',
2366 'channel': 'SciShow',
2367 'live_status': 'not_live',
2368 'duration': 248,
2369 'categories': ['Education'],
2370 'age_limit': 0,
2371 'channel_follower_count': int,
2372 'chapters': list,
2373 'uploader': 'SciShow',
2374 'uploader_url': 'https://www.youtube.com/@SciShow',
2375 'uploader_id': '@SciShow',
2376 'comment_count': int,
2377 'channel_is_verified': True,
2378 'heatmap': 'count:100',
2379 }, 'params': {'format': 'mhtml', 'skip_download': True}
2380 }, {
2381 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2382 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2383 'info_dict': {
2384 'id': '2NUZ8W2llS4',
2385 'ext': 'mp4',
2386 'title': 'The NP that test your phone performance 🙂',
2387 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2388 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2389 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2390 'duration': 21,
2391 'view_count': int,
2392 'age_limit': 0,
2393 'categories': ['Gaming'],
2394 'tags': 'count:23',
2395 'playable_in_embed': True,
2396 'live_status': 'not_live',
2397 'upload_date': '20220103',
2398 'like_count': int,
2399 'availability': 'public',
2400 'channel': 'Leon Nguyen',
2401 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2402 'comment_count': int,
2403 'channel_follower_count': int,
2404 'uploader': 'Leon Nguyen',
2405 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2406 'uploader_id': '@LeonNguyen',
2407 'heatmap': 'count:100',
2408 }
2409 }, {
2410 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2411 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2412 'info_dict': {
2413 'id': '2NUZ8W2llS4',
2414 'ext': 'mp4',
2415 'title': 'The NP that test your phone performance 🙂',
2416 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2417 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2418 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2419 'duration': 21,
2420 'view_count': int,
2421 'age_limit': 0,
2422 'categories': ['Gaming'],
2423 'tags': 'count:23',
2424 'playable_in_embed': True,
2425 'live_status': 'not_live',
2426 'upload_date': '20220102',
2427 'like_count': int,
2428 'availability': 'public',
2429 'channel': 'Leon Nguyen',
2430 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2431 'comment_count': int,
2432 'channel_follower_count': int,
2433 'uploader': 'Leon Nguyen',
2434 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2435 'uploader_id': '@LeonNguyen',
2436 'heatmap': 'count:100',
2437 },
2438 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2439 }, {
2440 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2441 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2442 'info_dict': {
2443 'id': 'mzZzzBU6lrM',
2444 'ext': 'mp4',
2445 'title': 'I Met GeorgeNotFound In Real Life...',
2446 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2447 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2448 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2449 'duration': 955,
2450 'view_count': int,
2451 'age_limit': 0,
2452 'categories': ['Entertainment'],
2453 'tags': 'count:26',
2454 'playable_in_embed': True,
2455 'live_status': 'not_live',
2456 'release_timestamp': 1641172509,
2457 'release_date': '20220103',
2458 'upload_date': '20220103',
2459 'like_count': int,
2460 'availability': 'public',
2461 'channel': 'Quackity',
2462 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2463 'channel_follower_count': int,
2464 'uploader': 'Quackity',
2465 'uploader_id': '@Quackity',
2466 'uploader_url': 'https://www.youtube.com/@Quackity',
2467 'comment_count': int,
2468 'channel_is_verified': True,
2469 'heatmap': 'count:100',
2470 }
2471 },
2472 { # continuous livestream. Microformat upload date should be preferred.
2473 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2474 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2475 'info_dict': {
2476 'id': 'kgx4WGK0oNU',
2477 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2478 'ext': 'mp4',
2479 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2480 'availability': 'public',
2481 'age_limit': 0,
2482 'release_timestamp': 1637975704,
2483 'upload_date': '20210619',
2484 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2485 'live_status': 'is_live',
2486 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2487 'channel': 'Abao in Tokyo',
2488 'channel_follower_count': int,
2489 'release_date': '20211127',
2490 'tags': 'count:39',
2491 'categories': ['People & Blogs'],
2492 'like_count': int,
2493 'view_count': int,
2494 'playable_in_embed': True,
2495 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2496 'concurrent_view_count': int,
2497 'uploader': 'Abao in Tokyo',
2498 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2499 'uploader_id': '@abaointokyo',
2500 },
2501 'params': {'skip_download': True}
2502 }, {
2503 # Story. Requires specific player params to work.
2504 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2505 'info_dict': {
2506 'id': 'vv8qTUWmulI',
2507 'ext': 'mp4',
2508 'availability': 'unlisted',
2509 'view_count': int,
2510 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2511 'upload_date': '20220526',
2512 'categories': ['Education'],
2513 'title': 'Story',
2514 'channel': 'IT\'S HISTORY',
2515 'description': '',
2516 'duration': 12,
2517 'playable_in_embed': True,
2518 'age_limit': 0,
2519 'live_status': 'not_live',
2520 'tags': [],
2521 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2522 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2523 },
2524 'skip': 'stories get removed after some period of time',
2525 }, {
2526 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2527 'info_dict': {
2528 'id': 'tjjjtzRLHvA',
2529 'ext': 'mp4',
2530 'title': 'ハッシュタグ無し };if window.ytcsi',
2531 'upload_date': '20220323',
2532 'like_count': int,
2533 'availability': 'unlisted',
2534 'channel': 'Lesmiscore',
2535 'thumbnail': r're:^https?://.*\.jpg',
2536 'age_limit': 0,
2537 'categories': ['Music'],
2538 'view_count': int,
2539 'description': '',
2540 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2541 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2542 'live_status': 'not_live',
2543 'playable_in_embed': True,
2544 'channel_follower_count': int,
2545 'duration': 6,
2546 'tags': [],
2547 'uploader_id': '@lesmiscore',
2548 'uploader': 'Lesmiscore',
2549 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2550 }
2551 }, {
2552 # Prefer primary title+description language metadata by default
2553 # Do not prefer translated description if primary is empty
2554 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2555 'info_dict': {
2556 'id': 'el3E4MbxRqQ',
2557 'ext': 'mp4',
2558 'title': 'dlp test video 2 - primary sv no desc',
2559 'description': '',
2560 'channel': 'cole-dlp-test-acc',
2561 'tags': [],
2562 'view_count': int,
2563 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2564 'like_count': int,
2565 'playable_in_embed': True,
2566 'availability': 'unlisted',
2567 'thumbnail': r're:^https?://.*\.jpg',
2568 'age_limit': 0,
2569 'duration': 5,
2570 'live_status': 'not_live',
2571 'upload_date': '20220908',
2572 'categories': ['People & Blogs'],
2573 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2574 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2575 'uploader_id': '@coletdjnz',
2576 'uploader': 'cole-dlp-test-acc',
2577 },
2578 'params': {'skip_download': True}
2579 }, {
2580 # Extractor argument: prefer translated title+description
2581 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2582 'info_dict': {
2583 'id': 'gHKT4uU8Zng',
2584 'ext': 'mp4',
2585 'channel': 'cole-dlp-test-acc',
2586 'tags': [],
2587 'duration': 5,
2588 'live_status': 'not_live',
2589 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2590 'upload_date': '20220728',
2591 'view_count': int,
2592 'categories': ['People & Blogs'],
2593 'thumbnail': r're:^https?://.*\.jpg',
2594 'title': 'dlp test video title translated (fr)',
2595 'availability': 'public',
2596 'age_limit': 0,
2597 'description': 'dlp test video description translated (fr)',
2598 'playable_in_embed': True,
2599 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2600 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2601 'uploader_id': '@coletdjnz',
2602 'uploader': 'cole-dlp-test-acc',
2603 },
2604 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2605 'expected_warnings': [r'Preferring "fr" translated fields'],
2606 }, {
2607 'note': '6 channel audio',
2608 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2609 'only_matching': True,
2610 }, {
2611 'note': 'Multiple HLS formats with same itag',
2612 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2613 'info_dict': {
2614 'id': 'kX3nB4PpJko',
2615 'ext': 'mp4',
2616 'categories': ['Entertainment'],
2617 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2618 'live_status': 'not_live',
2619 'duration': 937,
2620 'channel_follower_count': int,
2621 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2622 'title': 'Last To Take Hand Off Jet, Keeps It!',
2623 'channel': 'MrBeast',
2624 'playable_in_embed': True,
2625 'view_count': int,
2626 'upload_date': '20221112',
2627 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2628 'age_limit': 0,
2629 'availability': 'public',
2630 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2631 'like_count': int,
2632 'tags': [],
2633 'uploader': 'MrBeast',
2634 'uploader_url': 'https://www.youtube.com/@MrBeast',
2635 'uploader_id': '@MrBeast',
2636 'comment_count': int,
2637 'channel_is_verified': True,
2638 'heatmap': 'count:100',
2639 },
2640 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2641 }, {
2642 'note': 'Audio formats with Dynamic Range Compression',
2643 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2644 'info_dict': {
2645 'id': 'Tq92D6wQ1mg',
2646 'ext': 'webm',
2647 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2648 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2649 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2650 'channel_follower_count': int,
2651 'description': 'md5:17eccca93a786d51bc67646756894066',
2652 'upload_date': '20191228',
2653 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2654 'playable_in_embed': True,
2655 'like_count': int,
2656 'categories': ['Entertainment'],
2657 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2658 'age_limit': 18,
2659 'channel': 'Projekt Melody',
2660 'view_count': int,
2661 'availability': 'needs_auth',
2662 'comment_count': int,
2663 'live_status': 'not_live',
2664 'duration': 106,
2665 'uploader': 'Projekt Melody',
2666 'uploader_id': '@ProjektMelody',
2667 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2668 },
2669 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2670 },
2671 {
2672 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2673 'info_dict': {
2674 'id': 'qVv6vCqciTM',
2675 'ext': 'mp4',
2676 'age_limit': 0,
2677 'comment_count': int,
2678 'chapters': 'count:13',
2679 'upload_date': '20221223',
2680 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2681 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2682 'like_count': int,
2683 'release_date': '20221223',
2684 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2685 '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】',
2686 'view_count': int,
2687 'playable_in_embed': True,
2688 'duration': 4438,
2689 'availability': 'public',
2690 'channel_follower_count': int,
2691 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2692 'categories': ['Entertainment'],
2693 'live_status': 'was_live',
2694 'release_timestamp': 1671793345,
2695 'channel': 'さなちゃんねる',
2696 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2697 'uploader': 'さなちゃんねる',
2698 'uploader_url': 'https://www.youtube.com/@sana_natori',
2699 'uploader_id': '@sana_natori',
2700 'channel_is_verified': True,
2701 'heatmap': 'count:100',
2702 },
2703 },
2704 {
2705 # Fallbacks when webpage and web client is unavailable
2706 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2707 'info_dict': {
2708 'id': 'wSSmNUl9Snw',
2709 'ext': 'mp4',
2710 # 'categories': ['Science & Technology'],
2711 'view_count': int,
2712 'chapters': 'count:2',
2713 'channel': 'Scott Manley',
2714 'like_count': int,
2715 'age_limit': 0,
2716 # 'availability': 'public',
2717 'channel_follower_count': int,
2718 'live_status': 'not_live',
2719 'upload_date': '20170831',
2720 'duration': 682,
2721 'tags': 'count:8',
2722 'uploader_url': 'https://www.youtube.com/@scottmanley',
2723 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2724 'uploader': 'Scott Manley',
2725 'uploader_id': '@scottmanley',
2726 'title': 'The Computer Hack That Saved Apollo 14',
2727 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2728 'thumbnail': r're:^https?://.*\.webp',
2729 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2730 'playable_in_embed': True,
2731 'comment_count': int,
2732 'channel_is_verified': True,
2733 'heatmap': 'count:100',
2734 },
2735 'params': {
2736 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2737 },
2738 },
2739 ]
2740
2741 _WEBPAGE_TESTS = [
2742 # YouTube <object> embed
2743 {
2744 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2745 'md5': '873c81d308b979f0e23ee7e620b312a3',
2746 'info_dict': {
2747 'id': 'msN87y-iEx0',
2748 'ext': 'mp4',
2749 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2750 'upload_date': '20080526',
2751 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2752 'age_limit': 0,
2753 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2754 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2755 'playable_in_embed': True,
2756 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2757 'like_count': int,
2758 'comment_count': int,
2759 'channel': 'Christopher Sykes',
2760 'live_status': 'not_live',
2761 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2762 'availability': 'public',
2763 'duration': 195,
2764 'view_count': int,
2765 'categories': ['Science & Technology'],
2766 'channel_follower_count': int,
2767 'uploader': 'Christopher Sykes',
2768 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2769 'uploader_id': '@ChristopherSykesDocumentaries',
2770 'heatmap': 'count:100',
2771 },
2772 'params': {
2773 'skip_download': True,
2774 }
2775 },
2776 ]
2777
2778 @classmethod
2779 def suitable(cls, url):
2780 from ..utils import parse_qs
2781
2782 qs = parse_qs(url)
2783 if qs.get('list', [None])[0]:
2784 return False
2785 return super().suitable(url)
2786
2787 def __init__(self, *args, **kwargs):
2788 super().__init__(*args, **kwargs)
2789 self._code_cache = {}
2790 self._player_cache = {}
2791
2792 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2793 lock = threading.Lock()
2794 start_time = time.time()
2795 formats = [f for f in formats if f.get('is_from_start')]
2796
2797 def refetch_manifest(format_id, delay):
2798 nonlocal formats, start_time, is_live
2799 if time.time() <= start_time + delay:
2800 return
2801
2802 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2803 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2804 microformats = traverse_obj(
2805 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2806 expected_type=dict)
2807 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2808 is_live = live_status == 'is_live'
2809 start_time = time.time()
2810
2811 def mpd_feed(format_id, delay):
2812 """
2813 @returns (manifest_url, manifest_stream_number, is_live) or None
2814 """
2815 for retry in self.RetryManager(fatal=False):
2816 with lock:
2817 refetch_manifest(format_id, delay)
2818
2819 f = next((f for f in formats if f['format_id'] == format_id), None)
2820 if not f:
2821 if not is_live:
2822 retry.error = f'{video_id}: Video is no longer live'
2823 else:
2824 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2825 continue
2826 return f['manifest_url'], f['manifest_stream_number'], is_live
2827 return None
2828
2829 for f in formats:
2830 f['is_live'] = is_live
2831 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2832 live_start_time, mpd_feed, not is_live and f.copy())
2833 if is_live:
2834 f['fragments'] = gen
2835 f['protocol'] = 'http_dash_segments_generator'
2836 else:
2837 f['fragments'] = LazyList(gen({}))
2838 del f['is_from_start']
2839
2840 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2841 FETCH_SPAN, MAX_DURATION = 5, 432000
2842
2843 mpd_url, stream_number, is_live = None, None, True
2844
2845 begin_index = 0
2846 download_start_time = ctx.get('start') or time.time()
2847
2848 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2849 if lack_early_segments:
2850 self.report_warning(bug_reports_message(
2851 'Starting download from the last 120 hours of the live stream since '
2852 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2853 lack_early_segments = True
2854
2855 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2856 fragments, fragment_base_url = None, None
2857
2858 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2859 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2860 # Obtain from MPD's maximum seq value
2861 old_mpd_url = mpd_url
2862 last_error = ctx.pop('last_error', None)
2863 expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2864 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2865 or (mpd_url, stream_number, False))
2866 if not refresh_sequence:
2867 if expire_fast and not is_live:
2868 return False, last_seq
2869 elif old_mpd_url == mpd_url:
2870 return True, last_seq
2871 if manifestless_orig_fmt:
2872 fmt_info = manifestless_orig_fmt
2873 else:
2874 try:
2875 fmts, _ = self._extract_mpd_formats_and_subtitles(
2876 mpd_url, None, note=False, errnote=False, fatal=False)
2877 except ExtractorError:
2878 fmts = None
2879 if not fmts:
2880 no_fragment_score += 2
2881 return False, last_seq
2882 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2883 fragments = fmt_info['fragments']
2884 fragment_base_url = fmt_info['fragment_base_url']
2885 assert fragment_base_url
2886
2887 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2888 return True, _last_seq
2889
2890 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2891 while is_live:
2892 fetch_time = time.time()
2893 if no_fragment_score > 30:
2894 return
2895 if last_segment_url:
2896 # Obtain from "X-Head-Seqnum" header value from each segment
2897 try:
2898 urlh = self._request_webpage(
2899 last_segment_url, None, note=False, errnote=False, fatal=False)
2900 except ExtractorError:
2901 urlh = None
2902 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2903 if last_seq is None:
2904 no_fragment_score += 2
2905 last_segment_url = None
2906 continue
2907 else:
2908 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2909 no_fragment_score += 2
2910 if not should_continue:
2911 continue
2912
2913 if known_idx > last_seq:
2914 last_segment_url = None
2915 continue
2916
2917 last_seq += 1
2918
2919 if begin_index < 0 and known_idx < 0:
2920 # skip from the start when it's negative value
2921 known_idx = last_seq + begin_index
2922 if lack_early_segments:
2923 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2924 try:
2925 for idx in range(known_idx, last_seq):
2926 # do not update sequence here or you'll get skipped some part of it
2927 should_continue, _ = _extract_sequence_from_mpd(False, False)
2928 if not should_continue:
2929 known_idx = idx - 1
2930 raise ExtractorError('breaking out of outer loop')
2931 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2932 yield {
2933 'url': last_segment_url,
2934 'fragment_count': last_seq,
2935 }
2936 if known_idx == last_seq:
2937 no_fragment_score += 5
2938 else:
2939 no_fragment_score = 0
2940 known_idx = last_seq
2941 except ExtractorError:
2942 continue
2943
2944 if manifestless_orig_fmt:
2945 # Stop at the first iteration if running for post-live manifestless;
2946 # fragment count no longer increase since it starts
2947 break
2948
2949 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2950
2951 def _extract_player_url(self, *ytcfgs, webpage=None):
2952 player_url = traverse_obj(
2953 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2954 get_all=False, expected_type=str)
2955 if not player_url:
2956 return
2957 return urljoin('https://www.youtube.com', player_url)
2958
2959 def _download_player_url(self, video_id, fatal=False):
2960 res = self._download_webpage(
2961 'https://www.youtube.com/iframe_api',
2962 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2963 if res:
2964 player_version = self._search_regex(
2965 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2966 if player_version:
2967 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2968
2969 def _signature_cache_id(self, example_sig):
2970 """ Return a string representation of a signature """
2971 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2972
2973 @classmethod
2974 def _extract_player_info(cls, player_url):
2975 for player_re in cls._PLAYER_INFO_RE:
2976 id_m = re.search(player_re, player_url)
2977 if id_m:
2978 break
2979 else:
2980 raise ExtractorError('Cannot identify player %r' % player_url)
2981 return id_m.group('id')
2982
2983 def _load_player(self, video_id, player_url, fatal=True):
2984 player_id = self._extract_player_info(player_url)
2985 if player_id not in self._code_cache:
2986 code = self._download_webpage(
2987 player_url, video_id, fatal=fatal,
2988 note='Downloading player ' + player_id,
2989 errnote='Download of %s failed' % player_url)
2990 if code:
2991 self._code_cache[player_id] = code
2992 return self._code_cache.get(player_id)
2993
2994 def _extract_signature_function(self, video_id, player_url, example_sig):
2995 player_id = self._extract_player_info(player_url)
2996
2997 # Read from filesystem cache
2998 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2999 assert os.path.basename(func_id) == func_id
3000
3001 self.write_debug(f'Extracting signature function {func_id}')
3002 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
3003
3004 if not cache_spec:
3005 code = self._load_player(video_id, player_url)
3006 if code:
3007 res = self._parse_sig_js(code)
3008 test_string = ''.join(map(chr, range(len(example_sig))))
3009 cache_spec = [ord(c) for c in res(test_string)]
3010 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
3011
3012 return lambda s: ''.join(s[i] for i in cache_spec)
3013
3014 def _print_sig_code(self, func, example_sig):
3015 if not self.get_param('youtube_print_sig_code'):
3016 return
3017
3018 def gen_sig_code(idxs):
3019 def _genslice(start, end, step):
3020 starts = '' if start == 0 else str(start)
3021 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
3022 steps = '' if step == 1 else (':%d' % step)
3023 return f's[{starts}{ends}{steps}]'
3024
3025 step = None
3026 # Quelch pyflakes warnings - start will be set when step is set
3027 start = '(Never used)'
3028 for i, prev in zip(idxs[1:], idxs[:-1]):
3029 if step is not None:
3030 if i - prev == step:
3031 continue
3032 yield _genslice(start, prev, step)
3033 step = None
3034 continue
3035 if i - prev in [-1, 1]:
3036 step = i - prev
3037 start = prev
3038 continue
3039 else:
3040 yield 's[%d]' % prev
3041 if step is None:
3042 yield 's[%d]' % i
3043 else:
3044 yield _genslice(start, i, step)
3045
3046 test_string = ''.join(map(chr, range(len(example_sig))))
3047 cache_res = func(test_string)
3048 cache_spec = [ord(c) for c in cache_res]
3049 expr_code = ' + '.join(gen_sig_code(cache_spec))
3050 signature_id_tuple = '(%s)' % (
3051 ', '.join(str(len(p)) for p in example_sig.split('.')))
3052 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3053 ' return %s\n') % (signature_id_tuple, expr_code)
3054 self.to_screen('Extracted signature function:\n' + code)
3055
3056 def _parse_sig_js(self, jscode):
3057 funcname = self._search_regex(
3058 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3059 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3060 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3061 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3062 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}\(a,\d+\))?',
3063 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3064 # Obsolete patterns
3065 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3066 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3067 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3068 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3069 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3070 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3071 jscode, 'Initial JS player signature function name', group='sig')
3072
3073 jsi = JSInterpreter(jscode)
3074 initial_function = jsi.extract_function(funcname)
3075 return lambda s: initial_function([s])
3076
3077 def _cached(self, func, *cache_id):
3078 def inner(*args, **kwargs):
3079 if cache_id not in self._player_cache:
3080 try:
3081 self._player_cache[cache_id] = func(*args, **kwargs)
3082 except ExtractorError as e:
3083 self._player_cache[cache_id] = e
3084 except Exception as e:
3085 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3086
3087 ret = self._player_cache[cache_id]
3088 if isinstance(ret, Exception):
3089 raise ret
3090 return ret
3091 return inner
3092
3093 def _decrypt_signature(self, s, video_id, player_url):
3094 """Turn the encrypted s field into a working signature"""
3095 extract_sig = self._cached(
3096 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3097 func = extract_sig(video_id, player_url, s)
3098 self._print_sig_code(func, s)
3099 return func(s)
3100
3101 def _decrypt_nsig(self, s, video_id, player_url):
3102 """Turn the encrypted n field into a working signature"""
3103 if player_url is None:
3104 raise ExtractorError('Cannot decrypt nsig without player_url')
3105 player_url = urljoin('https://www.youtube.com', player_url)
3106
3107 try:
3108 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3109 except ExtractorError as e:
3110 raise ExtractorError('Unable to extract nsig function code', cause=e)
3111 if self.get_param('youtube_print_sig_code'):
3112 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3113
3114 try:
3115 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3116 ret = extract_nsig(jsi, func_code)(s)
3117 except JSInterpreter.Exception as e:
3118 try:
3119 jsi = PhantomJSwrapper(self, timeout=5000)
3120 except ExtractorError:
3121 raise e
3122 self.report_warning(
3123 f'Native nsig extraction failed: Trying with PhantomJS\n'
3124 f' n = {s} ; player = {player_url}', video_id)
3125 self.write_debug(e, only_once=True)
3126
3127 args, func_body = func_code
3128 ret = jsi.execute(
3129 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3130 video_id=video_id, note='Executing signature code').strip()
3131
3132 self.write_debug(f'Decrypted nsig {s} => {ret}')
3133 return ret
3134
3135 def _extract_n_function_name(self, jscode):
3136 funcname, idx = self._search_regex(
3137 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3138 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3139 if not idx:
3140 return funcname
3141
3142 return json.loads(js_to_json(self._search_regex(
3143 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])[,;]', jscode,
3144 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3145
3146 def _extract_n_function_code(self, video_id, player_url):
3147 player_id = self._extract_player_info(player_url)
3148 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3149 jscode = func_code or self._load_player(video_id, player_url)
3150 jsi = JSInterpreter(jscode)
3151
3152 if func_code:
3153 return jsi, player_id, func_code
3154
3155 func_name = self._extract_n_function_name(jscode)
3156
3157 # For redundancy
3158 func_code = self._search_regex(
3159 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3160 # NB: The end of the regex is intentionally kept strict
3161 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3162 jscode, 'nsig function', group=('var', 'code'), default=None)
3163 if func_code:
3164 func_code = ([func_code[0]], func_code[1])
3165 else:
3166 self.write_debug('Extracting nsig function with jsinterp')
3167 func_code = jsi.extract_function_code(func_name)
3168
3169 self.cache.store('youtube-nsig', player_id, func_code)
3170 return jsi, player_id, func_code
3171
3172 def _extract_n_function_from_code(self, jsi, func_code):
3173 func = jsi.extract_function_from_code(*func_code)
3174
3175 def extract_nsig(s):
3176 try:
3177 ret = func([s])
3178 except JSInterpreter.Exception:
3179 raise
3180 except Exception as e:
3181 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3182
3183 if ret.startswith('enhanced_except_'):
3184 raise JSInterpreter.Exception('Signature function returned an exception')
3185 return ret
3186
3187 return extract_nsig
3188
3189 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3190 """
3191 Extract signatureTimestamp (sts)
3192 Required to tell API what sig/player version is in use.
3193 """
3194 sts = None
3195 if isinstance(ytcfg, dict):
3196 sts = int_or_none(ytcfg.get('STS'))
3197
3198 if not sts:
3199 # Attempt to extract from player
3200 if player_url is None:
3201 error_msg = 'Cannot extract signature timestamp without player_url.'
3202 if fatal:
3203 raise ExtractorError(error_msg)
3204 self.report_warning(error_msg)
3205 return
3206 code = self._load_player(video_id, player_url, fatal=fatal)
3207 if code:
3208 sts = int_or_none(self._search_regex(
3209 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3210 'JS player signature timestamp', group='sts', fatal=fatal))
3211 return sts
3212
3213 def _mark_watched(self, video_id, player_responses):
3214 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3215 label = 'fully ' if is_full else ''
3216 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3217 expected_type=url_or_none)
3218 if not url:
3219 self.report_warning(f'Unable to mark {label}watched')
3220 return
3221 parsed_url = urllib.parse.urlparse(url)
3222 qs = urllib.parse.parse_qs(parsed_url.query)
3223
3224 # cpn generation algorithm is reverse engineered from base.js.
3225 # In fact it works even with dummy cpn.
3226 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3227 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3228
3229 # # more consistent results setting it to right before the end
3230 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3231
3232 qs.update({
3233 'ver': ['2'],
3234 'cpn': [cpn],
3235 'cmt': video_length,
3236 'el': 'detailpage', # otherwise defaults to "shorts"
3237 })
3238
3239 if is_full:
3240 # these seem to mark watchtime "history" in the real world
3241 # they're required, so send in a single value
3242 qs.update({
3243 'st': 0,
3244 'et': video_length,
3245 })
3246
3247 url = urllib.parse.urlunparse(
3248 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3249
3250 self._download_webpage(
3251 url, video_id, f'Marking {label}watched',
3252 'Unable to mark watched', fatal=False)
3253
3254 @classmethod
3255 def _extract_from_webpage(cls, url, webpage):
3256 # Invidious Instances
3257 # https://github.com/yt-dlp/yt-dlp/issues/195
3258 # https://github.com/iv-org/invidious/pull/1730
3259 mobj = re.search(
3260 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3261 webpage)
3262 if mobj:
3263 yield cls.url_result(mobj.group('url'), cls)
3264 raise cls.StopExtraction()
3265
3266 yield from super()._extract_from_webpage(url, webpage)
3267
3268 # lazyYT YouTube embed
3269 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3270 yield cls.url_result(unescapeHTML(id_), cls, id_)
3271
3272 # Wordpress "YouTube Video Importer" plugin
3273 for m in re.findall(r'''(?x)<div[^>]+
3274 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3275 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3276 yield cls.url_result(m[-1], cls, m[-1])
3277
3278 @classmethod
3279 def extract_id(cls, url):
3280 video_id = cls.get_temp_id(url)
3281 if not video_id:
3282 raise ExtractorError(f'Invalid URL: {url}')
3283 return video_id
3284
3285 def _extract_chapters_from_json(self, data, duration):
3286 chapter_list = traverse_obj(
3287 data, (
3288 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3289 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3290 ), expected_type=list)
3291
3292 return self._extract_chapters_helper(
3293 chapter_list,
3294 start_function=lambda chapter: float_or_none(
3295 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3296 title_function=lambda chapter: traverse_obj(
3297 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3298 duration=duration)
3299
3300 def _extract_chapters_from_engagement_panel(self, data, duration):
3301 content_list = traverse_obj(
3302 data,
3303 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3304 expected_type=list)
3305 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3306 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3307
3308 return next(filter(None, (
3309 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3310 chapter_time, chapter_title, duration)
3311 for contents in content_list)), [])
3312
3313 def _extract_heatmap_from_player_overlay(self, data):
3314 content_list = traverse_obj(data, (
3315 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer', 'decoratedPlayerBarRenderer', 'playerBar',
3316 'multiMarkersPlayerBarRenderer', 'markersMap', ..., 'value', 'heatmap', 'heatmapRenderer', 'heatMarkers', {list}))
3317 return next(filter(None, (
3318 traverse_obj(contents, (..., 'heatMarkerRenderer', {
3319 'start_time': ('timeRangeStartMillis', {functools.partial(float_or_none, scale=1000)}),
3320 'end_time': {lambda x: (x['timeRangeStartMillis'] + x['markerDurationMillis']) / 1000},
3321 'value': ('heatMarkerIntensityScoreNormalized', {float_or_none}),
3322 })) for contents in content_list)), None)
3323
3324 def _extract_comment(self, comment_renderer, parent=None):
3325 comment_id = comment_renderer.get('commentId')
3326 if not comment_id:
3327 return
3328
3329 info = {
3330 'id': comment_id,
3331 'text': self._get_text(comment_renderer, 'contentText'),
3332 'like_count': self._get_count(comment_renderer, 'voteCount'),
3333 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3334 'author': self._get_text(comment_renderer, 'authorText'),
3335 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3336 'parent': parent or 'root',
3337 }
3338
3339 # Timestamp is an estimate calculated from the current time and time_text
3340 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3341 timestamp = self._parse_time_text(time_text)
3342
3343 info.update({
3344 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3345 '_time_text': time_text,
3346 'timestamp': timestamp,
3347 })
3348
3349 info['author_url'] = urljoin(
3350 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3351 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3352 expected_type=str, get_all=False))
3353
3354 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3355 if author_is_uploader is not None:
3356 info['author_is_uploader'] = author_is_uploader
3357
3358 comment_abr = traverse_obj(
3359 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3360 if comment_abr is not None:
3361 info['is_favorited'] = 'creatorHeart' in comment_abr
3362
3363 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3364 if self._has_badge(badges, BadgeType.VERIFIED):
3365 info['author_is_verified'] = True
3366
3367 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3368 if is_pinned:
3369 info['is_pinned'] = True
3370
3371 return info
3372
3373 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3374
3375 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3376
3377 def extract_header(contents):
3378 _continuation = None
3379 for content in contents:
3380 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3381 expected_comment_count = self._get_count(
3382 comments_header_renderer, 'countText', 'commentsCount')
3383
3384 if expected_comment_count is not None:
3385 tracker['est_total'] = expected_comment_count
3386 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3387 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3388
3389 sort_menu_item = try_get(
3390 comments_header_renderer,
3391 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3392 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3393
3394 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3395 if not _continuation:
3396 continue
3397
3398 sort_text = str_or_none(sort_menu_item.get('title'))
3399 if not sort_text:
3400 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3401 self.to_screen('Sorting comments by %s' % sort_text.lower())
3402 break
3403 return _continuation
3404
3405 def extract_thread(contents):
3406 if not parent:
3407 tracker['current_page_thread'] = 0
3408 for content in contents:
3409 if not parent and tracker['total_parent_comments'] >= max_parents:
3410 yield
3411 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3412 comment_renderer = get_first(
3413 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3414 expected_type=dict, default={})
3415
3416 comment = self._extract_comment(comment_renderer, parent)
3417 if not comment:
3418 continue
3419 comment_id = comment['id']
3420 if comment.get('is_pinned'):
3421 tracker['pinned_comment_ids'].add(comment_id)
3422 # Sometimes YouTube may break and give us infinite looping comments.
3423 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3424 if comment_id in tracker['seen_comment_ids']:
3425 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3426 # Pinned comments may appear a second time in newest first sort
3427 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3428 continue
3429 self.report_warning('Detected YouTube comments looping. Stopping comment extraction as we probably cannot get any more.')
3430 yield
3431 else:
3432 tracker['seen_comment_ids'].add(comment['id'])
3433
3434 tracker['running_total'] += 1
3435 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3436 yield comment
3437
3438 # Attempt to get the replies
3439 comment_replies_renderer = try_get(
3440 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3441
3442 if comment_replies_renderer:
3443 tracker['current_page_thread'] += 1
3444 comment_entries_iter = self._comment_entries(
3445 comment_replies_renderer, ytcfg, video_id,
3446 parent=comment.get('id'), tracker=tracker)
3447 yield from itertools.islice(comment_entries_iter, min(
3448 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3449
3450 # Keeps track of counts across recursive calls
3451 if not tracker:
3452 tracker = dict(
3453 running_total=0,
3454 est_total=None,
3455 current_page_thread=0,
3456 total_parent_comments=0,
3457 total_reply_comments=0,
3458 seen_comment_ids=set(),
3459 pinned_comment_ids=set()
3460 )
3461
3462 # TODO: Deprecated
3463 # YouTube comments have a max depth of 2
3464 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3465 if max_depth:
3466 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3467 'Set max replies in the max-comments extractor argument instead')
3468 if max_depth == 1 and parent:
3469 return
3470
3471 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3472 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3473
3474 continuation = self._extract_continuation(root_continuation_data)
3475
3476 response = None
3477 is_forced_continuation = False
3478 is_first_continuation = parent is None
3479 if is_first_continuation and not continuation:
3480 # Sometimes you can get comments by generating the continuation yourself,
3481 # even if YouTube initially reports them being disabled - e.g. stories comments.
3482 # Note: if the comment section is actually disabled, YouTube may return a response with
3483 # required check_get_keys missing. So we will disable that check initially in this case.
3484 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3485 is_forced_continuation = True
3486
3487 continuation_items_path = (
3488 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3489 for page_num in itertools.count(0):
3490 if not continuation:
3491 break
3492 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3493 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3494 if page_num == 0:
3495 if is_first_continuation:
3496 note_prefix = 'Downloading comment section API JSON'
3497 else:
3498 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3499 tracker['current_page_thread'], comment_prog_str)
3500 else:
3501 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3502 ' ' if parent else '', ' replies' if parent else '',
3503 page_num, comment_prog_str)
3504
3505 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3506 # Ignore check if YouTube says the comment count is 0.
3507 check_get_keys = None
3508 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3509 check_get_keys = [[*continuation_items_path, ..., (
3510 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3511 try:
3512 response = self._extract_response(
3513 item_id=None, query=continuation,
3514 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3515 check_get_keys=check_get_keys)
3516 except ExtractorError as e:
3517 # Ignore incomplete data error for replies if retries didn't work.
3518 # This is to allow any other parent comments and comment threads to be downloaded.
3519 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3520 if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
3521 self.report_warning(
3522 'Received incomplete data for a comment reply thread and retrying did not help. '
3523 'Ignoring to let other comments be downloaded.')
3524 else:
3525 raise
3526 is_forced_continuation = False
3527 continuation = None
3528 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3529 if is_first_continuation:
3530 continuation = extract_header(continuation_items)
3531 is_first_continuation = False
3532 if continuation:
3533 break
3534 continue
3535
3536 for entry in extract_thread(continuation_items):
3537 if not entry:
3538 return
3539 yield entry
3540 continuation = self._extract_continuation({'contents': continuation_items})
3541 if continuation:
3542 break
3543
3544 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3545 if message and not parent and tracker['running_total'] == 0:
3546 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3547 raise self.CommentsDisabled
3548
3549 @staticmethod
3550 def _generate_comment_continuation(video_id):
3551 """
3552 Generates initial comment section continuation token from given video id
3553 """
3554 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3555 return base64.b64encode(token.encode()).decode()
3556
3557 def _get_comments(self, ytcfg, video_id, contents, webpage):
3558 """Entry for comment extraction"""
3559 def _real_comment_extract(contents):
3560 renderer = next((
3561 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3562 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3563 yield from self._comment_entries(renderer, ytcfg, video_id)
3564
3565 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3566 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3567
3568 @staticmethod
3569 def _get_checkok_params():
3570 return {'contentCheckOk': True, 'racyCheckOk': True}
3571
3572 @classmethod
3573 def _generate_player_context(cls, sts=None):
3574 context = {
3575 'html5Preference': 'HTML5_PREF_WANTS',
3576 }
3577 if sts is not None:
3578 context['signatureTimestamp'] = sts
3579 return {
3580 'playbackContext': {
3581 'contentPlaybackContext': context
3582 },
3583 **cls._get_checkok_params()
3584 }
3585
3586 @staticmethod
3587 def _is_agegated(player_response):
3588 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3589 return True
3590
3591 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3592 AGE_GATE_REASONS = (
3593 'confirm your age', 'age-restricted', 'inappropriate', # reason
3594 'age_verification_required', 'age_check_required', # status
3595 )
3596 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3597
3598 @staticmethod
3599 def _is_unplayable(player_response):
3600 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3601
3602 _PLAYER_PARAMS = 'CgIQBg=='
3603
3604 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3605
3606 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3607 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3608 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3609 headers = self.generate_api_headers(
3610 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3611
3612 yt_query = {
3613 'videoId': video_id,
3614 }
3615 if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
3616 yt_query['params'] = self._PLAYER_PARAMS
3617
3618 yt_query.update(self._generate_player_context(sts))
3619 return self._extract_response(
3620 item_id=video_id, ep='player', query=yt_query,
3621 ytcfg=player_ytcfg, headers=headers, fatal=True,
3622 default_client=client,
3623 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3624 ) or None
3625
3626 def _get_requested_clients(self, url, smuggled_data):
3627 requested_clients = []
3628 default = ['ios', 'android', 'web']
3629 allowed_clients = sorted(
3630 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3631 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3632 for client in self._configuration_arg('player_client'):
3633 if client in allowed_clients:
3634 requested_clients.append(client)
3635 elif client == 'default':
3636 requested_clients.extend(default)
3637 elif client == 'all':
3638 requested_clients.extend(allowed_clients)
3639 else:
3640 self.report_warning(f'Skipping unsupported client {client}')
3641 if not requested_clients:
3642 requested_clients = default
3643
3644 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3645 requested_clients.extend(
3646 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3647
3648 return orderedSet(requested_clients)
3649
3650 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3651 initial_pr = None
3652 if webpage:
3653 initial_pr = self._search_json(
3654 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3655
3656 all_clients = set(clients)
3657 clients = clients[::-1]
3658 prs = []
3659
3660 def append_client(*client_names):
3661 """ Append the first client name that exists but not already used """
3662 for client_name in client_names:
3663 actual_client = _split_innertube_client(client_name)[0]
3664 if actual_client in INNERTUBE_CLIENTS:
3665 if actual_client not in all_clients:
3666 clients.append(client_name)
3667 all_clients.add(actual_client)
3668 return
3669
3670 # Android player_response does not have microFormats which are needed for
3671 # extraction of some data. So we return the initial_pr with formats
3672 # stripped out even if not requested by the user
3673 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3674 if initial_pr:
3675 pr = dict(initial_pr)
3676 pr['streamingData'] = None
3677 prs.append(pr)
3678
3679 last_error = None
3680 tried_iframe_fallback = False
3681 player_url = None
3682 while clients:
3683 client, base_client, variant = _split_innertube_client(clients.pop())
3684 player_ytcfg = master_ytcfg if client == 'web' else {}
3685 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3686 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3687
3688 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3689 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3690 if 'js' in self._configuration_arg('player_skip'):
3691 require_js_player = False
3692 player_url = None
3693
3694 if not player_url and not tried_iframe_fallback and require_js_player:
3695 player_url = self._download_player_url(video_id)
3696 tried_iframe_fallback = True
3697
3698 try:
3699 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3700 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3701 except ExtractorError as e:
3702 if last_error:
3703 self.report_warning(last_error)
3704 last_error = e
3705 continue
3706
3707 if pr:
3708 # YouTube may return a different video player response than expected.
3709 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3710 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3711 if pr_video_id and pr_video_id != video_id:
3712 self.report_warning(
3713 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3714 else:
3715 # Save client name for introspection later
3716 name = short_client_name(client)
3717 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3718 sd[STREAMING_DATA_CLIENT_NAME] = name
3719 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3720 f[STREAMING_DATA_CLIENT_NAME] = name
3721 prs.append(pr)
3722
3723 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3724 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3725 append_client(f'{base_client}_creator')
3726 elif self._is_agegated(pr):
3727 if variant == 'tv_embedded':
3728 append_client(f'{base_client}_embedded')
3729 elif not variant:
3730 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3731
3732 if last_error:
3733 if not len(prs):
3734 raise last_error
3735 self.report_warning(last_error)
3736 return prs, player_url
3737
3738 def _needs_live_processing(self, live_status, duration):
3739 if (live_status == 'is_live' and self.get_param('live_from_start')
3740 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3741 return live_status
3742
3743 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3744 CHUNK_SIZE = 10 << 20
3745 itags, stream_ids = collections.defaultdict(set), []
3746 itag_qualities, res_qualities = {}, {0: None}
3747 q = qualities([
3748 # Normally tiny is the smallest video-only formats. But
3749 # audio-only formats with unknown quality may get tagged as tiny
3750 'tiny',
3751 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3752 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3753 ])
3754 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3755 format_types = self._configuration_arg('formats')
3756 all_formats = 'duplicate' in format_types
3757 if self._configuration_arg('include_duplicate_formats'):
3758 all_formats = True
3759 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3760 'Use formats=duplicate extractor argument instead')
3761
3762 def build_fragments(f):
3763 return LazyList({
3764 'url': update_url_query(f['url'], {
3765 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3766 })
3767 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3768
3769 for fmt in streaming_formats:
3770 if fmt.get('targetDurationSec'):
3771 continue
3772
3773 itag = str_or_none(fmt.get('itag'))
3774 audio_track = fmt.get('audioTrack') or {}
3775 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3776 if not all_formats:
3777 if stream_id in stream_ids:
3778 continue
3779
3780 quality = fmt.get('quality')
3781 height = int_or_none(fmt.get('height'))
3782 if quality == 'tiny' or not quality:
3783 quality = fmt.get('audioQuality', '').lower() or quality
3784 # The 3gp format (17) in android client has a quality of "small",
3785 # but is actually worse than other formats
3786 if itag == '17':
3787 quality = 'tiny'
3788 if quality:
3789 if itag:
3790 itag_qualities[itag] = quality
3791 if height:
3792 res_qualities[height] = quality
3793 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3794 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3795 # number of fragment that would subsequently requested with (`&sq=N`)
3796 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3797 continue
3798
3799 fmt_url = fmt.get('url')
3800 if not fmt_url:
3801 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3802 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3803 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3804 if not all((sc, fmt_url, player_url, encrypted_sig)):
3805 continue
3806 try:
3807 fmt_url += '&%s=%s' % (
3808 traverse_obj(sc, ('sp', -1)) or 'signature',
3809 self._decrypt_signature(encrypted_sig, video_id, player_url)
3810 )
3811 except ExtractorError as e:
3812 self.report_warning('Signature extraction failed: Some formats may be missing',
3813 video_id=video_id, only_once=True)
3814 self.write_debug(e, only_once=True)
3815 continue
3816
3817 query = parse_qs(fmt_url)
3818 throttled = False
3819 if query.get('n'):
3820 try:
3821 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3822 fmt_url = update_url_query(fmt_url, {
3823 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3824 })
3825 except ExtractorError as e:
3826 phantomjs_hint = ''
3827 if isinstance(e, JSInterpreter.Exception):
3828 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3829 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3830 if player_url:
3831 self.report_warning(
3832 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3833 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3834 self.write_debug(e, only_once=True)
3835 else:
3836 self.report_warning(
3837 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3838 video_id=video_id, only_once=True)
3839 throttled = True
3840
3841 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3842 language_preference = (
3843 10 if audio_track.get('audioIsDefault') and 10
3844 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3845 else -1)
3846 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3847 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3848 # Make sure to avoid false positives with small duration differences.
3849 # E.g. __2ABJjxzNo, ySuUZEjARPY
3850 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3851 if is_damaged:
3852 self.report_warning(
3853 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3854
3855 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3856 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3857 fps = int_or_none(fmt.get('fps')) or 0
3858 dct = {
3859 'asr': int_or_none(fmt.get('audioSampleRate')),
3860 'filesize': int_or_none(fmt.get('contentLength')),
3861 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3862 'format_note': join_nonempty(
3863 join_nonempty(audio_track.get('displayName'),
3864 language_preference > 0 and ' (default)', delim=''),
3865 name, fmt.get('isDrc') and 'DRC',
3866 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3867 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3868 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3869 (self.get_param('verbose') or all_formats) and client_name,
3870 delim=', '),
3871 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3872 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3873 + (100 if 'Premium' in name else 0)),
3874 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3875 'audio_channels': fmt.get('audioChannels'),
3876 'height': height,
3877 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3878 'has_drm': bool(fmt.get('drmFamilies')),
3879 'tbr': tbr,
3880 'url': fmt_url,
3881 'width': int_or_none(fmt.get('width')),
3882 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3883 'desc' if language_preference < -1 else '') or None,
3884 'language_preference': language_preference,
3885 # Strictly de-prioritize damaged and 3gp formats
3886 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3887 }
3888 mime_mobj = re.match(
3889 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3890 if mime_mobj:
3891 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3892 dct.update(parse_codecs(mime_mobj.group(2)))
3893 if itag:
3894 itags[itag].add(('https', dct.get('language')))
3895 stream_ids.append(stream_id)
3896 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3897 if single_stream and dct.get('ext'):
3898 dct['container'] = dct['ext'] + '_dash'
3899
3900 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3901 yield {
3902 **dct,
3903 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3904 'protocol': 'http_dash_segments',
3905 'fragments': build_fragments(dct),
3906 }
3907 if all_formats or 'dashy' not in format_types:
3908 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3909 yield dct
3910
3911 needs_live_processing = self._needs_live_processing(live_status, duration)
3912 skip_bad_formats = 'incomplete' not in format_types
3913 if self._configuration_arg('include_incomplete_formats'):
3914 skip_bad_formats = False
3915 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3916 'Use formats=incomplete extractor argument instead')
3917
3918 skip_manifests = set(self._configuration_arg('skip'))
3919 if (not self.get_param('youtube_include_hls_manifest', True)
3920 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3921 or needs_live_processing and skip_bad_formats):
3922 skip_manifests.add('hls')
3923
3924 if not self.get_param('youtube_include_dash_manifest', True):
3925 skip_manifests.add('dash')
3926 if self._configuration_arg('include_live_dash'):
3927 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3928 'Use formats=incomplete extractor argument instead')
3929 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3930 skip_manifests.add('dash')
3931
3932 def process_manifest_format(f, proto, client_name, itag):
3933 key = (proto, f.get('language'))
3934 if not all_formats and key in itags[itag]:
3935 return False
3936 itags[itag].add(key)
3937
3938 if itag and all_formats:
3939 f['format_id'] = f'{itag}-{proto}'
3940 elif any(p != proto for p, _ in itags[itag]):
3941 f['format_id'] = f'{itag}-{proto}'
3942 elif itag:
3943 f['format_id'] = itag
3944
3945 if itag in ('616', '235'):
3946 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
3947 f['source_preference'] = (f.get('source_preference') or -1) + 100
3948
3949 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3950 if f['quality'] == -1 and f.get('height'):
3951 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3952 if self.get_param('verbose') or all_formats:
3953 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3954 if f.get('fps') and f['fps'] <= 1:
3955 del f['fps']
3956 return True
3957
3958 subtitles = {}
3959 for sd in streaming_data:
3960 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3961
3962 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3963 if hls_manifest_url:
3964 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3965 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3966 subtitles = self._merge_subtitles(subs, subtitles)
3967 for f in fmts:
3968 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3969 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3970 yield f
3971
3972 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3973 if dash_manifest_url:
3974 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3975 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3976 for f in formats:
3977 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3978 f['filesize'] = int_or_none(self._search_regex(
3979 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3980 if needs_live_processing:
3981 f['is_from_start'] = True
3982
3983 yield f
3984 yield subtitles
3985
3986 def _extract_storyboard(self, player_responses, duration):
3987 spec = get_first(
3988 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3989 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3990 if not base_url:
3991 return
3992 L = len(spec) - 1
3993 for i, args in enumerate(spec):
3994 args = args.split('#')
3995 counts = list(map(int_or_none, args[:5]))
3996 if len(args) != 8 or not all(counts):
3997 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3998 continue
3999 width, height, frame_count, cols, rows = counts
4000 N, sigh = args[6:]
4001
4002 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
4003 fragment_count = frame_count / (cols * rows)
4004 fragment_duration = duration / fragment_count
4005 yield {
4006 'format_id': f'sb{i}',
4007 'format_note': 'storyboard',
4008 'ext': 'mhtml',
4009 'protocol': 'mhtml',
4010 'acodec': 'none',
4011 'vcodec': 'none',
4012 'url': url,
4013 'width': width,
4014 'height': height,
4015 'fps': frame_count / duration,
4016 'rows': rows,
4017 'columns': cols,
4018 'fragments': [{
4019 'url': url.replace('$M', str(j)),
4020 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4021 } for j in range(math.ceil(fragment_count))],
4022 }
4023
4024 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4025 webpage = None
4026 if 'webpage' not in self._configuration_arg('player_skip'):
4027 query = {'bpctr': '9999999999', 'has_verified': '1'}
4028 if smuggled_data.get('is_story'): # XXX: Deprecated
4029 query['pp'] = self._PLAYER_PARAMS
4030 webpage = self._download_webpage(
4031 webpage_url, video_id, fatal=False, query=query)
4032
4033 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4034
4035 player_responses, player_url = self._extract_player_responses(
4036 self._get_requested_clients(url, smuggled_data),
4037 video_id, webpage, master_ytcfg, smuggled_data)
4038
4039 return webpage, master_ytcfg, player_responses, player_url
4040
4041 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4042 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4043 is_live = get_first(video_details, 'isLive')
4044 if is_live is None:
4045 is_live = get_first(live_broadcast_details, 'isLiveNow')
4046 live_content = get_first(video_details, 'isLiveContent')
4047 is_upcoming = get_first(video_details, 'isUpcoming')
4048 post_live = get_first(video_details, 'isPostLiveDvr')
4049 live_status = ('post_live' if post_live
4050 else 'is_live' if is_live
4051 else 'is_upcoming' if is_upcoming
4052 else 'was_live' if live_content
4053 else 'not_live' if False in (is_live, live_content)
4054 else None)
4055 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4056 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4057
4058 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4059
4060 def _real_extract(self, url):
4061 url, smuggled_data = unsmuggle_url(url, {})
4062 video_id = self._match_id(url)
4063
4064 base_url = self.http_scheme() + '//www.youtube.com/'
4065 webpage_url = base_url + 'watch?v=' + video_id
4066
4067 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4068
4069 playability_statuses = traverse_obj(
4070 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4071
4072 trailer_video_id = get_first(
4073 playability_statuses,
4074 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4075 expected_type=str)
4076 if trailer_video_id:
4077 return self.url_result(
4078 trailer_video_id, self.ie_key(), trailer_video_id)
4079
4080 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4081 if webpage else (lambda x: None))
4082
4083 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4084 microformats = traverse_obj(
4085 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4086 expected_type=dict)
4087
4088 translated_title = self._get_text(microformats, (..., 'title'))
4089 video_title = (self._preferred_lang and translated_title
4090 or get_first(video_details, 'title') # primary
4091 or translated_title
4092 or search_meta(['og:title', 'twitter:title', 'title']))
4093 translated_description = self._get_text(microformats, (..., 'description'))
4094 original_description = get_first(video_details, 'shortDescription')
4095 video_description = (
4096 self._preferred_lang and translated_description
4097 # If original description is blank, it will be an empty string.
4098 # Do not prefer translated description in this case.
4099 or original_description if original_description is not None else translated_description)
4100
4101 multifeed_metadata_list = get_first(
4102 player_responses,
4103 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4104 expected_type=str)
4105 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4106 if self.get_param('noplaylist'):
4107 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4108 else:
4109 entries = []
4110 feed_ids = []
4111 for feed in multifeed_metadata_list.split(','):
4112 # Unquote should take place before split on comma (,) since textual
4113 # fields may contain comma as well (see
4114 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4115 feed_data = urllib.parse.parse_qs(
4116 urllib.parse.unquote_plus(feed))
4117
4118 def feed_entry(name):
4119 return try_get(
4120 feed_data, lambda x: x[name][0], str)
4121
4122 feed_id = feed_entry('id')
4123 if not feed_id:
4124 continue
4125 feed_title = feed_entry('title')
4126 title = video_title
4127 if feed_title:
4128 title += ' (%s)' % feed_title
4129 entries.append({
4130 '_type': 'url_transparent',
4131 'ie_key': 'Youtube',
4132 'url': smuggle_url(
4133 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4134 {'force_singlefeed': True}),
4135 'title': title,
4136 })
4137 feed_ids.append(feed_id)
4138 self.to_screen(
4139 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4140 % (', '.join(feed_ids), video_id))
4141 return self.playlist_result(
4142 entries, video_id, video_title, video_description)
4143
4144 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4145 or int_or_none(get_first(microformats, 'lengthSeconds'))
4146 or parse_duration(search_meta('duration')) or None)
4147
4148 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4149 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4150 if live_status == 'post_live':
4151 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4152
4153 if not formats:
4154 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4155 self.report_drm(video_id)
4156 pemr = get_first(
4157 playability_statuses,
4158 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4159 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4160 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4161 if subreason:
4162 if subreason == 'The uploader has not made this video available in your country.':
4163 countries = get_first(microformats, 'availableCountries')
4164 if not countries:
4165 regions_allowed = search_meta('regionsAllowed')
4166 countries = regions_allowed.split(',') if regions_allowed else None
4167 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4168 reason += f'. {subreason}'
4169 if reason:
4170 self.raise_no_formats(reason, expected=True)
4171
4172 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4173 if not keywords and webpage:
4174 keywords = [
4175 unescapeHTML(m.group('content'))
4176 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4177 for keyword in keywords:
4178 if keyword.startswith('yt:stretch='):
4179 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4180 if mobj:
4181 # NB: float is intentional for forcing float division
4182 w, h = (float(v) for v in mobj.groups())
4183 if w > 0 and h > 0:
4184 ratio = w / h
4185 for f in formats:
4186 if f.get('vcodec') != 'none':
4187 f['stretched_ratio'] = ratio
4188 break
4189 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4190 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4191 if thumbnail_url:
4192 thumbnails.append({
4193 'url': thumbnail_url,
4194 })
4195 original_thumbnails = thumbnails.copy()
4196
4197 # The best resolution thumbnails sometimes does not appear in the webpage
4198 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4199 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4200 thumbnail_names = [
4201 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4202 # in resolution, these are not the custom thumbnail. So de-prioritize them
4203 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4204 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4205 ]
4206 n_thumbnail_names = len(thumbnail_names)
4207 thumbnails.extend({
4208 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4209 video_id=video_id, name=name, ext=ext,
4210 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4211 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4212 for thumb in thumbnails:
4213 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4214 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4215 self._remove_duplicate_formats(thumbnails)
4216 self._downloader._sort_thumbnails(original_thumbnails)
4217
4218 category = get_first(microformats, 'category') or search_meta('genre')
4219 channel_id = self.ucid_or_none(str_or_none(
4220 get_first(video_details, 'channelId')
4221 or get_first(microformats, 'externalChannelId')
4222 or search_meta('channelId')))
4223 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4224
4225 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4226 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4227 if not duration and live_end_time and live_start_time:
4228 duration = live_end_time - live_start_time
4229
4230 needs_live_processing = self._needs_live_processing(live_status, duration)
4231
4232 def is_bad_format(fmt):
4233 if needs_live_processing and not fmt.get('is_from_start'):
4234 return True
4235 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4236 and fmt.get('protocol') == 'http_dash_segments'):
4237 return True
4238
4239 for fmt in filter(is_bad_format, formats):
4240 fmt['preference'] = (fmt.get('preference') or -1) - 10
4241 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4242
4243 if needs_live_processing:
4244 self._prepare_live_from_start_formats(
4245 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4246
4247 formats.extend(self._extract_storyboard(player_responses, duration))
4248
4249 channel_handle = self.handle_from_url(owner_profile_url)
4250
4251 info = {
4252 'id': video_id,
4253 'title': video_title,
4254 'formats': formats,
4255 'thumbnails': thumbnails,
4256 # The best thumbnail that we are sure exists. Prevents unnecessary
4257 # URL checking if user don't care about getting the best possible thumbnail
4258 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4259 'description': video_description,
4260 'channel_id': channel_id,
4261 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4262 'duration': duration,
4263 'view_count': int_or_none(
4264 get_first((video_details, microformats), (..., 'viewCount'))
4265 or search_meta('interactionCount')),
4266 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4267 'age_limit': 18 if (
4268 get_first(microformats, 'isFamilySafe') is False
4269 or search_meta('isFamilyFriendly') == 'false'
4270 or search_meta('og:restrictions:age') == '18+') else 0,
4271 'webpage_url': webpage_url,
4272 'categories': [category] if category else None,
4273 'tags': keywords,
4274 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4275 'live_status': live_status,
4276 'release_timestamp': live_start_time,
4277 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4278 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4279 }
4280
4281 subtitles = {}
4282 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4283 if pctr:
4284 def get_lang_code(track):
4285 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4286 or track.get('languageCode'))
4287
4288 # Converted into dicts to remove duplicates
4289 captions = {
4290 get_lang_code(sub): sub
4291 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4292 translation_languages = {
4293 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4294 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4295
4296 def process_language(container, base_url, lang_code, sub_name, query):
4297 lang_subs = container.setdefault(lang_code, [])
4298 for fmt in self._SUBTITLE_FORMATS:
4299 query.update({
4300 'fmt': fmt,
4301 })
4302 lang_subs.append({
4303 'ext': fmt,
4304 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4305 'name': sub_name,
4306 })
4307
4308 # NB: Constructing the full subtitle dictionary is slow
4309 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4310 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4311 for lang_code, caption_track in captions.items():
4312 base_url = caption_track.get('baseUrl')
4313 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4314 if not base_url:
4315 continue
4316 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4317 if caption_track.get('kind') != 'asr':
4318 if not lang_code:
4319 continue
4320 process_language(
4321 subtitles, base_url, lang_code, lang_name, {})
4322 if not caption_track.get('isTranslatable'):
4323 continue
4324 for trans_code, trans_name in translation_languages.items():
4325 if not trans_code:
4326 continue
4327 orig_trans_code = trans_code
4328 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4329 if not get_translated_subs:
4330 continue
4331 trans_code += f'-{lang_code}'
4332 trans_name += format_field(lang_name, None, ' from %s')
4333 if lang_code == f'a-{orig_trans_code}':
4334 # Set audio language based on original subtitles
4335 for f in formats:
4336 if f.get('acodec') != 'none' and not f.get('language'):
4337 f['language'] = orig_trans_code
4338 # Add an "-orig" label to the original language so that it can be distinguished.
4339 # The subs are returned without "-orig" as well for compatibility
4340 process_language(
4341 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4342 # Setting tlang=lang returns damaged subtitles.
4343 process_language(automatic_captions, base_url, trans_code, trans_name,
4344 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4345
4346 info['automatic_captions'] = automatic_captions
4347 info['subtitles'] = subtitles
4348
4349 parsed_url = urllib.parse.urlparse(url)
4350 for component in [parsed_url.fragment, parsed_url.query]:
4351 query = urllib.parse.parse_qs(component)
4352 for k, v in query.items():
4353 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4354 d_k += '_time'
4355 if d_k not in info and k in s_ks:
4356 info[d_k] = parse_duration(query[k][0])
4357
4358 # Youtube Music Auto-generated description
4359 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4360 # XXX: Causes catastrophic backtracking if description has "·"
4361 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4362 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4363 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4364 mobj = re.search(
4365 r'''(?xs)
4366 (?=(?P<track>[^\n·]+))(?P=track)·
4367 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4368 (?=(?P<album>[^\n]+))(?P=album)\n
4369 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4370 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4371 (.+?\nArtist\s*:\s*
4372 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4373 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4374 ''', video_description)
4375 if mobj:
4376 release_year = mobj.group('release_year')
4377 release_date = mobj.group('release_date')
4378 if release_date:
4379 release_date = release_date.replace('-', '')
4380 if not release_year:
4381 release_year = release_date[:4]
4382 info.update({
4383 'album': mobj.group('album'.strip()),
4384 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4385 'track': mobj.group('track').strip(),
4386 'release_date': release_date,
4387 'release_year': int_or_none(release_year),
4388 })
4389
4390 initial_data = None
4391 if webpage:
4392 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4393 if not traverse_obj(initial_data, 'contents'):
4394 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4395 initial_data = None
4396 if not initial_data:
4397 query = {'videoId': video_id}
4398 query.update(self._get_checkok_params())
4399 initial_data = self._extract_response(
4400 item_id=video_id, ep='next', fatal=False,
4401 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4402 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4403 note='Downloading initial data API JSON')
4404
4405 info['comment_count'] = traverse_obj(initial_data, (
4406 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4407 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4408 ), (
4409 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4410 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4411 ), expected_type=self._get_count, get_all=False)
4412
4413 try: # This will error if there is no livechat
4414 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4415 except (KeyError, IndexError, TypeError):
4416 pass
4417 else:
4418 info.setdefault('subtitles', {})['live_chat'] = [{
4419 # url is needed to set cookies
4420 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4421 'video_id': video_id,
4422 'ext': 'json',
4423 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4424 else 'youtube_live_chat_replay'),
4425 }]
4426
4427 if initial_data:
4428 info['chapters'] = (
4429 self._extract_chapters_from_json(initial_data, duration)
4430 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4431 or self._extract_chapters_from_description(video_description, duration)
4432 or None)
4433
4434 info['heatmap'] = self._extract_heatmap_from_player_overlay(initial_data)
4435
4436 contents = traverse_obj(
4437 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4438 expected_type=list, default=[])
4439
4440 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4441 if vpir:
4442 stl = vpir.get('superTitleLink')
4443 if stl:
4444 stl = self._get_text(stl)
4445 if try_get(
4446 vpir,
4447 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4448 info['location'] = stl
4449 else:
4450 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4451 if mobj:
4452 info.update({
4453 'series': mobj.group(1),
4454 'season_number': int(mobj.group(2)),
4455 'episode_number': int(mobj.group(3)),
4456 })
4457 for tlb in (try_get(
4458 vpir,
4459 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4460 list) or []):
4461 tbrs = variadic(
4462 traverse_obj(
4463 tlb, ('toggleButtonRenderer', ...),
4464 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4465 for tbr in tbrs:
4466 for getter, regex in [(
4467 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4468 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4469 lambda x: x['accessibility'],
4470 lambda x: x['accessibilityData']['accessibilityData'],
4471 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4472 label = (try_get(tbr, getter, dict) or {}).get('label')
4473 if label:
4474 mobj = re.match(regex, label)
4475 if mobj:
4476 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4477 break
4478 sbr_tooltip = try_get(
4479 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4480 if sbr_tooltip:
4481 like_count, dislike_count = sbr_tooltip.split(' / ')
4482 info.update({
4483 'like_count': str_to_int(like_count),
4484 'dislike_count': str_to_int(dislike_count),
4485 })
4486 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4487 if vcr:
4488 vc = self._get_count(vcr, 'viewCount')
4489 # Upcoming premieres with waiting count are treated as live here
4490 if vcr.get('isLive'):
4491 info['concurrent_view_count'] = vc
4492 elif info.get('view_count') is None:
4493 info['view_count'] = vc
4494
4495 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4496 if vsir:
4497 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4498 info.update({
4499 'channel': self._get_text(vor, 'title'),
4500 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4501
4502 if not channel_handle:
4503 channel_handle = self.handle_from_url(
4504 traverse_obj(vor, (
4505 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4506 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4507 {str}), get_all=False))
4508
4509 rows = try_get(
4510 vsir,
4511 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4512 list) or []
4513 multiple_songs = False
4514 for row in rows:
4515 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4516 multiple_songs = True
4517 break
4518 for row in rows:
4519 mrr = row.get('metadataRowRenderer') or {}
4520 mrr_title = mrr.get('title')
4521 if not mrr_title:
4522 continue
4523 mrr_title = self._get_text(mrr, 'title')
4524 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4525 if mrr_title == 'License':
4526 info['license'] = mrr_contents_text
4527 elif not multiple_songs:
4528 if mrr_title == 'Album':
4529 info['album'] = mrr_contents_text
4530 elif mrr_title == 'Artist':
4531 info['artist'] = mrr_contents_text
4532 elif mrr_title == 'Song':
4533 info['track'] = mrr_contents_text
4534 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4535 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4536 info['channel_is_verified'] = True
4537
4538 info.update({
4539 'uploader': info.get('channel'),
4540 'uploader_id': channel_handle,
4541 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4542 })
4543 # The upload date for scheduled, live and past live streams / premieres in microformats
4544 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4545 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4546 upload_date = (
4547 unified_strdate(get_first(microformats, 'uploadDate'))
4548 or unified_strdate(search_meta('uploadDate')))
4549 if not upload_date or (
4550 live_status in ('not_live', None)
4551 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4552 ):
4553 upload_date = strftime_or_none(
4554 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4555 info['upload_date'] = upload_date
4556
4557 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4558 v = info.get(s_k)
4559 if v:
4560 info[d_k] = v
4561
4562 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4563
4564 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4565 or get_first(video_details, 'isPrivate', expected_type=bool))
4566
4567 info['availability'] = (
4568 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4569 else self._availability(
4570 is_private=is_private,
4571 needs_premium=(
4572 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4573 or False if initial_data and is_private is not None else None),
4574 needs_subscription=(
4575 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4576 or False if initial_data and is_private is not None else None),
4577 needs_auth=info['age_limit'] >= 18,
4578 is_unlisted=None if is_private is None else (
4579 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4580 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4581
4582 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4583
4584 self.mark_watched(video_id, player_responses)
4585
4586 return info
4587
4588
4589 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4590 @staticmethod
4591 def passthrough_smuggled_data(func):
4592 def _smuggle(info, smuggled_data):
4593 if info.get('_type') not in ('url', 'url_transparent'):
4594 return info
4595 if smuggled_data.get('is_music_url'):
4596 parsed_url = urllib.parse.urlparse(info['url'])
4597 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4598 smuggled_data.pop('is_music_url')
4599 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4600 if smuggled_data:
4601 info['url'] = smuggle_url(info['url'], smuggled_data)
4602 return info
4603
4604 @functools.wraps(func)
4605 def wrapper(self, url):
4606 url, smuggled_data = unsmuggle_url(url, {})
4607 if self.is_music_url(url):
4608 smuggled_data['is_music_url'] = True
4609 info_dict = func(self, url, smuggled_data)
4610 if smuggled_data:
4611 _smuggle(info_dict, smuggled_data)
4612 if info_dict.get('entries'):
4613 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4614 return info_dict
4615 return wrapper
4616
4617 @staticmethod
4618 def _extract_basic_item_renderer(item):
4619 # Modified from _extract_grid_item_renderer
4620 known_basic_renderers = (
4621 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4622 )
4623 for key, renderer in item.items():
4624 if not isinstance(renderer, dict):
4625 continue
4626 elif key in known_basic_renderers:
4627 return renderer
4628 elif key.startswith('grid') and key.endswith('Renderer'):
4629 return renderer
4630
4631 def _extract_channel_renderer(self, renderer):
4632 channel_id = self.ucid_or_none(renderer['channelId'])
4633 title = self._get_text(renderer, 'title')
4634 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4635 channel_handle = self.handle_from_url(
4636 traverse_obj(renderer, (
4637 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4638 ('browseEndpoint', 'canonicalBaseUrl')),
4639 {str}), get_all=False))
4640 if not channel_handle:
4641 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4642 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4643 return {
4644 '_type': 'url',
4645 'url': channel_url,
4646 'id': channel_id,
4647 'ie_key': YoutubeTabIE.ie_key(),
4648 'channel': title,
4649 'uploader': title,
4650 'channel_id': channel_id,
4651 'channel_url': channel_url,
4652 'title': title,
4653 'uploader_id': channel_handle,
4654 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4655 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4656 # However, in feed/channels this is set correctly to the subscriber count
4657 'channel_follower_count': traverse_obj(
4658 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4659 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4660 'playlist_count': (
4661 # videoCountText may be the subscriber count
4662 self._get_count(renderer, 'videoCountText')
4663 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4664 'description': self._get_text(renderer, 'descriptionSnippet'),
4665 'channel_is_verified': True if self._has_badge(
4666 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4667 }
4668
4669 def _grid_entries(self, grid_renderer):
4670 for item in grid_renderer['items']:
4671 if not isinstance(item, dict):
4672 continue
4673 renderer = self._extract_basic_item_renderer(item)
4674 if not isinstance(renderer, dict):
4675 continue
4676 title = self._get_text(renderer, 'title')
4677
4678 # playlist
4679 playlist_id = renderer.get('playlistId')
4680 if playlist_id:
4681 yield self.url_result(
4682 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4683 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4684 video_title=title)
4685 continue
4686 # video
4687 video_id = renderer.get('videoId')
4688 if video_id:
4689 yield self._extract_video(renderer)
4690 continue
4691 # channel
4692 channel_id = renderer.get('channelId')
4693 if channel_id:
4694 yield self._extract_channel_renderer(renderer)
4695 continue
4696 # generic endpoint URL support
4697 ep_url = urljoin('https://www.youtube.com/', try_get(
4698 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4699 str))
4700 if ep_url:
4701 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4702 if ie.suitable(ep_url):
4703 yield self.url_result(
4704 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4705 break
4706
4707 def _music_reponsive_list_entry(self, renderer):
4708 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4709 if video_id:
4710 title = traverse_obj(renderer, (
4711 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4712 'text', 'runs', 0, 'text'))
4713 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4714 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4715 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4716 if playlist_id:
4717 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4718 if video_id:
4719 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4720 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4721 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4722 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4723 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4724 if browse_id:
4725 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4726 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4727
4728 def _shelf_entries_from_content(self, shelf_renderer):
4729 content = shelf_renderer.get('content')
4730 if not isinstance(content, dict):
4731 return
4732 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4733 if renderer:
4734 # TODO: add support for nested playlists so each shelf is processed
4735 # as separate playlist
4736 # TODO: this includes only first N items
4737 yield from self._grid_entries(renderer)
4738 renderer = content.get('horizontalListRenderer')
4739 if renderer:
4740 # TODO
4741 pass
4742
4743 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4744 ep = try_get(
4745 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4746 str)
4747 shelf_url = urljoin('https://www.youtube.com', ep)
4748 if shelf_url:
4749 # Skipping links to another channels, note that checking for
4750 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4751 # will not work
4752 if skip_channels and '/channels?' in shelf_url:
4753 return
4754 title = self._get_text(shelf_renderer, 'title')
4755 yield self.url_result(shelf_url, video_title=title)
4756 # Shelf may not contain shelf URL, fallback to extraction from content
4757 yield from self._shelf_entries_from_content(shelf_renderer)
4758
4759 def _playlist_entries(self, video_list_renderer):
4760 for content in video_list_renderer['contents']:
4761 if not isinstance(content, dict):
4762 continue
4763 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4764 if not isinstance(renderer, dict):
4765 continue
4766 video_id = renderer.get('videoId')
4767 if not video_id:
4768 continue
4769 yield self._extract_video(renderer)
4770
4771 def _rich_entries(self, rich_grid_renderer):
4772 renderer = traverse_obj(
4773 rich_grid_renderer,
4774 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4775 video_id = renderer.get('videoId')
4776 if video_id:
4777 yield self._extract_video(renderer)
4778 return
4779 playlist_id = renderer.get('playlistId')
4780 if playlist_id:
4781 yield self.url_result(
4782 f'https://www.youtube.com/playlist?list={playlist_id}',
4783 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4784 video_title=self._get_text(renderer, 'title'))
4785 return
4786
4787 def _video_entry(self, video_renderer):
4788 video_id = video_renderer.get('videoId')
4789 if video_id:
4790 return self._extract_video(video_renderer)
4791
4792 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4793 url = urljoin('https://youtube.com', traverse_obj(
4794 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4795 if url:
4796 return self.url_result(
4797 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4798
4799 def _post_thread_entries(self, post_thread_renderer):
4800 post_renderer = try_get(
4801 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4802 if not post_renderer:
4803 return
4804 # video attachment
4805 video_renderer = try_get(
4806 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4807 video_id = video_renderer.get('videoId')
4808 if video_id:
4809 entry = self._extract_video(video_renderer)
4810 if entry:
4811 yield entry
4812 # playlist attachment
4813 playlist_id = try_get(
4814 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4815 if playlist_id:
4816 yield self.url_result(
4817 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4818 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4819 # inline video links
4820 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4821 for run in runs:
4822 if not isinstance(run, dict):
4823 continue
4824 ep_url = try_get(
4825 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4826 if not ep_url:
4827 continue
4828 if not YoutubeIE.suitable(ep_url):
4829 continue
4830 ep_video_id = YoutubeIE._match_id(ep_url)
4831 if video_id == ep_video_id:
4832 continue
4833 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4834
4835 def _post_thread_continuation_entries(self, post_thread_continuation):
4836 contents = post_thread_continuation.get('contents')
4837 if not isinstance(contents, list):
4838 return
4839 for content in contents:
4840 renderer = content.get('backstagePostThreadRenderer')
4841 if isinstance(renderer, dict):
4842 yield from self._post_thread_entries(renderer)
4843 continue
4844 renderer = content.get('videoRenderer')
4845 if isinstance(renderer, dict):
4846 yield self._video_entry(renderer)
4847
4848 r''' # unused
4849 def _rich_grid_entries(self, contents):
4850 for content in contents:
4851 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4852 if video_renderer:
4853 entry = self._video_entry(video_renderer)
4854 if entry:
4855 yield entry
4856 '''
4857
4858 def _report_history_entries(self, renderer):
4859 for url in traverse_obj(renderer, (
4860 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4861 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4862 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4863 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4864
4865 def _extract_entries(self, parent_renderer, continuation_list):
4866 # continuation_list is modified in-place with continuation_list = [continuation_token]
4867 continuation_list[:] = [None]
4868 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4869 for content in contents:
4870 if not isinstance(content, dict):
4871 continue
4872 is_renderer = traverse_obj(
4873 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4874 expected_type=dict)
4875 if not is_renderer:
4876 if content.get('richItemRenderer'):
4877 for entry in self._rich_entries(content['richItemRenderer']):
4878 yield entry
4879 continuation_list[0] = self._extract_continuation(parent_renderer)
4880 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4881 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4882 yield from self._report_history_entries(table)
4883 continuation_list[0] = self._extract_continuation(table)
4884 continue
4885
4886 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4887 for isr_content in isr_contents:
4888 if not isinstance(isr_content, dict):
4889 continue
4890
4891 known_renderers = {
4892 'playlistVideoListRenderer': self._playlist_entries,
4893 'gridRenderer': self._grid_entries,
4894 'reelShelfRenderer': self._grid_entries,
4895 'shelfRenderer': self._shelf_entries,
4896 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4897 'backstagePostThreadRenderer': self._post_thread_entries,
4898 'videoRenderer': lambda x: [self._video_entry(x)],
4899 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4900 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4901 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4902 }
4903 for key, renderer in isr_content.items():
4904 if key not in known_renderers:
4905 continue
4906 for entry in known_renderers[key](renderer):
4907 if entry:
4908 yield entry
4909 continuation_list[0] = self._extract_continuation(renderer)
4910 break
4911
4912 if not continuation_list[0]:
4913 continuation_list[0] = self._extract_continuation(is_renderer)
4914
4915 if not continuation_list[0]:
4916 continuation_list[0] = self._extract_continuation(parent_renderer)
4917
4918 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4919 continuation_list = [None]
4920 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4921 tab_content = try_get(tab, lambda x: x['content'], dict)
4922 if not tab_content:
4923 return
4924 parent_renderer = (
4925 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4926 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4927 yield from extract_entries(parent_renderer)
4928 continuation = continuation_list[0]
4929
4930 for page_num in itertools.count(1):
4931 if not continuation:
4932 break
4933 headers = self.generate_api_headers(
4934 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4935 response = self._extract_response(
4936 item_id=f'{item_id} page {page_num}',
4937 query=continuation, headers=headers, ytcfg=ytcfg,
4938 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4939
4940 if not response:
4941 break
4942 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4943 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4944 visitor_data = self._extract_visitor_data(response) or visitor_data
4945
4946 known_renderers = {
4947 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4948 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4949 'gridVideoRenderer': (self._grid_entries, 'items'),
4950 'gridChannelRenderer': (self._grid_entries, 'items'),
4951 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4952 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4953 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4954 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4955 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4956 'playlistVideoListContinuation': (self._playlist_entries, None),
4957 'gridContinuation': (self._grid_entries, None),
4958 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4959 'sectionListContinuation': (extract_entries, None), # for feeds
4960 }
4961
4962 continuation_items = traverse_obj(response, (
4963 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4964 'appendContinuationItemsAction', 'continuationItems'
4965 ), 'continuationContents', get_all=False)
4966 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4967
4968 video_items_renderer = None
4969 for key in continuation_item.keys():
4970 if key not in known_renderers:
4971 continue
4972 func, parent_key = known_renderers[key]
4973 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4974 continuation_list = [None]
4975 yield from func(video_items_renderer)
4976 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4977
4978 if not video_items_renderer:
4979 break
4980
4981 @staticmethod
4982 def _extract_selected_tab(tabs, fatal=True):
4983 for tab_renderer in tabs:
4984 if tab_renderer.get('selected'):
4985 return tab_renderer
4986 if fatal:
4987 raise ExtractorError('Unable to find selected tab')
4988
4989 @staticmethod
4990 def _extract_tab_renderers(response):
4991 return traverse_obj(
4992 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4993
4994 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4995 metadata = self._extract_metadata_from_tabs(item_id, data)
4996
4997 selected_tab = self._extract_selected_tab(tabs)
4998 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4999 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
5000
5001 return self.playlist_result(
5002 self._entries(
5003 selected_tab, metadata['id'], ytcfg,
5004 self._extract_account_syncid(ytcfg, data),
5005 self._extract_visitor_data(data, ytcfg)),
5006 **metadata)
5007
5008 def _extract_metadata_from_tabs(self, item_id, data):
5009 info = {'id': item_id}
5010
5011 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5012 if metadata_renderer:
5013 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5014 ('channelUrl', {self.ucid_from_url}))
5015 info.update({
5016 'channel': metadata_renderer.get('title'),
5017 'channel_id': channel_id,
5018 })
5019 if info['channel_id']:
5020 info['id'] = info['channel_id']
5021 else:
5022 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5023
5024 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5025 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5026 def _get_uncropped(url):
5027 return url_or_none((url or '').split('=')[0] + '=s0')
5028
5029 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5030 if avatar_thumbnails:
5031 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5032 if uncropped_avatar:
5033 avatar_thumbnails.append({
5034 'url': uncropped_avatar,
5035 'id': 'avatar_uncropped',
5036 'preference': 1
5037 })
5038
5039 channel_banners = self._extract_thumbnails(
5040 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5041 for banner in channel_banners:
5042 banner['preference'] = -10
5043
5044 if channel_banners:
5045 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5046 if uncropped_banner:
5047 channel_banners.append({
5048 'url': uncropped_banner,
5049 'id': 'banner_uncropped',
5050 'preference': -5
5051 })
5052
5053 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5054 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5055 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5056
5057 primary_thumbnails = self._extract_thumbnails(
5058 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5059 playlist_thumbnails = self._extract_thumbnails(
5060 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5061
5062 info.update({
5063 'title': (traverse_obj(metadata_renderer, 'title')
5064 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5065 or info['id']),
5066 'availability': self._extract_availability(data),
5067 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
5068 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5069 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
5070 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5071 })
5072
5073 channel_handle = (
5074 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5075 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5076
5077 if channel_handle:
5078 info.update({
5079 'uploader_id': channel_handle,
5080 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5081 })
5082
5083 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5084 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5085 info['channel_is_verified'] = True
5086 # Playlist stats is a text runs array containing [video count, view count, last updated].
5087 # last updated or (view count and last updated) may be missing.
5088 playlist_stats = get_first(
5089 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5090
5091 last_updated_unix = self._parse_time_text(
5092 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5093 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5094 info['modified_date'] = strftime_or_none(last_updated_unix)
5095
5096 info['view_count'] = self._get_count(playlist_stats, 1)
5097 if info['view_count'] is None: # 0 is allowed
5098 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5099 if info['view_count'] is None:
5100 info['view_count'] = self._get_count(data, (
5101 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5102 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5103
5104 info['playlist_count'] = self._get_count(playlist_stats, 0)
5105 if info['playlist_count'] is None: # 0 is allowed
5106 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5107
5108 if not info.get('channel_id'):
5109 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5110 if not owner: # Deprecated
5111 owner = traverse_obj(
5112 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5113 ('videoOwner', 'videoOwnerRenderer', 'title'))
5114 owner_text = self._get_text(owner)
5115 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5116 info.update({
5117 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5118 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5119 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5120 })
5121
5122 info.update({
5123 'uploader': info['channel'],
5124 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5125 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5126 })
5127
5128 return info
5129
5130 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5131 first_id = last_id = response = None
5132 for page_num in itertools.count(1):
5133 videos = list(self._playlist_entries(playlist))
5134 if not videos:
5135 return
5136 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5137 if start >= len(videos):
5138 return
5139 yield from videos[start:]
5140 first_id = first_id or videos[0]['id']
5141 last_id = videos[-1]['id']
5142 watch_endpoint = try_get(
5143 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5144 headers = self.generate_api_headers(
5145 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5146 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5147 query = {
5148 'playlistId': playlist_id,
5149 'videoId': watch_endpoint.get('videoId') or last_id,
5150 'index': watch_endpoint.get('index') or len(videos),
5151 'params': watch_endpoint.get('params') or 'OAE%3D'
5152 }
5153 response = self._extract_response(
5154 item_id='%s page %d' % (playlist_id, page_num),
5155 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5156 check_get_keys='contents'
5157 )
5158 playlist = try_get(
5159 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5160
5161 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5162 title = playlist.get('title') or try_get(
5163 data, lambda x: x['titleText']['simpleText'], str)
5164 playlist_id = playlist.get('playlistId') or item_id
5165
5166 # Delegating everything except mix playlists to regular tab-based playlist URL
5167 playlist_url = urljoin(url, try_get(
5168 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5169 str))
5170
5171 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5172 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5173 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5174
5175 if playlist_url and playlist_url != url and not is_known_unviewable:
5176 return self.url_result(
5177 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5178 video_title=title)
5179
5180 return self.playlist_result(
5181 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5182 playlist_id=playlist_id, playlist_title=title)
5183
5184 def _extract_availability(self, data):
5185 """
5186 Gets the availability of a given playlist/tab.
5187 Note: Unless YouTube tells us explicitly, we do not assume it is public
5188 @param data: response
5189 """
5190 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5191 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5192 player_header_privacy = playlist_header_renderer.get('privacy')
5193
5194 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5195
5196 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5197 privacy_setting_icon = get_first(
5198 (playlist_header_renderer, sidebar_renderer),
5199 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5200 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5201 expected_type=str)
5202
5203 microformats_is_unlisted = traverse_obj(
5204 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5205
5206 return (
5207 'public' if (
5208 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5209 or player_header_privacy == 'PUBLIC'
5210 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5211 else self._availability(
5212 is_private=(
5213 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5214 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5215 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5216 is_unlisted=(
5217 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5218 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5219 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5220 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5221 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5222 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5223 needs_auth=False))
5224
5225 @staticmethod
5226 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5227 sidebar_renderer = try_get(
5228 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5229 for item in sidebar_renderer:
5230 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5231 if renderer:
5232 return renderer
5233
5234 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5235 """
5236 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5237 """
5238 is_playlist = bool(traverse_obj(
5239 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5240 if not is_playlist:
5241 return
5242 headers = self.generate_api_headers(
5243 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5244 visitor_data=self._extract_visitor_data(data, ytcfg))
5245 query = {
5246 'params': 'wgYCCAA=',
5247 'browseId': f'VL{item_id}'
5248 }
5249 return self._extract_response(
5250 item_id=item_id, headers=headers, query=query,
5251 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5252 note='Redownloading playlist API JSON with unavailable videos')
5253
5254 @functools.cached_property
5255 def skip_webpage(self):
5256 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5257
5258 def _extract_webpage(self, url, item_id, fatal=True):
5259 webpage, data = None, None
5260 for retry in self.RetryManager(fatal=fatal):
5261 try:
5262 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5263 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5264 except ExtractorError as e:
5265 if isinstance(e.cause, network_exceptions):
5266 if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
5267 retry.error = e
5268 continue
5269 self._error_or_warning(e, fatal=fatal)
5270 break
5271
5272 try:
5273 self._extract_and_report_alerts(data)
5274 except ExtractorError as e:
5275 self._error_or_warning(e, fatal=fatal)
5276 break
5277
5278 # Sometimes youtube returns a webpage with incomplete ytInitialData
5279 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5280 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5281 retry.error = ExtractorError('Incomplete yt initial data received')
5282 continue
5283
5284 return webpage, data
5285
5286 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5287 """Use if failed to extract ytcfg (and data) from initial webpage"""
5288 if not ytcfg and self.is_authenticated:
5289 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5290 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5291 raise ExtractorError(
5292 f'{msg}. If you are not downloading private content, or '
5293 'your cookies are only for the first account and channel,'
5294 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5295 expected=True)
5296 self.report_warning(msg, only_once=True)
5297
5298 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5299 data = None
5300 if not self.skip_webpage:
5301 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5302 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5303 # Reject webpage data if redirected to home page without explicitly requesting
5304 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5305 if (url != 'https://www.youtube.com/feed/recommended'
5306 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5307 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5308 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5309 if fatal:
5310 raise ExtractorError(msg, expected=True)
5311 self.report_warning(msg, only_once=True)
5312 if not data:
5313 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5314 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5315 return data, ytcfg
5316
5317 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5318 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5319 resolve_response = self._extract_response(
5320 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5321 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5322 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5323 for ep_key, ep in endpoints.items():
5324 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5325 if params:
5326 return self._extract_response(
5327 item_id=item_id, query=params, ep=ep, headers=headers,
5328 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5329 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5330 err_note = 'Failed to resolve url (does the playlist exist?)'
5331 if fatal:
5332 raise ExtractorError(err_note, expected=True)
5333 self.report_warning(err_note, item_id)
5334
5335 _SEARCH_PARAMS = None
5336
5337 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5338 data = {'query': query}
5339 if params is NO_DEFAULT:
5340 params = self._SEARCH_PARAMS
5341 if params:
5342 data['params'] = params
5343
5344 content_keys = (
5345 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5346 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5347 # ytmusic search
5348 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5349 ('continuationContents', ),
5350 )
5351 display_id = f'query "{query}"'
5352 check_get_keys = tuple({keys[0] for keys in content_keys})
5353 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5354 self._report_playlist_authcheck(ytcfg, fatal=False)
5355
5356 continuation_list = [None]
5357 search = None
5358 for page_num in itertools.count(1):
5359 data.update(continuation_list[0] or {})
5360 headers = self.generate_api_headers(
5361 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5362 search = self._extract_response(
5363 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5364 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5365 slr_contents = traverse_obj(search, *content_keys)
5366 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5367 if not continuation_list[0]:
5368 break
5369
5370
5371 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5372 IE_DESC = 'YouTube Tabs'
5373 _VALID_URL = r'''(?x:
5374 https?://
5375 (?!consent\.)(?:\w+\.)?
5376 (?:
5377 youtube(?:kids)?\.com|
5378 %(invidious)s
5379 )/
5380 (?:
5381 (?P<channel_type>channel|c|user|browse)/|
5382 (?P<not_channel>
5383 feed/|hashtag/|
5384 (?:playlist|watch)\?.*?\blist=
5385 )|
5386 (?!(?:%(reserved_names)s)\b) # Direct URLs
5387 )
5388 (?P<id>[^/?\#&]+)
5389 )''' % {
5390 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5391 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5392 }
5393 IE_NAME = 'youtube:tab'
5394
5395 _TESTS = [{
5396 'note': 'playlists, multipage',
5397 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5398 'playlist_mincount': 94,
5399 'info_dict': {
5400 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5401 'title': 'Igor Kleiner - Playlists',
5402 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5403 'uploader': 'Igor Kleiner',
5404 'uploader_id': '@IgorDataScience',
5405 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5406 'channel': 'Igor Kleiner',
5407 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5408 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5409 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5410 'channel_follower_count': int
5411 },
5412 }, {
5413 'note': 'playlists, multipage, different order',
5414 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5415 'playlist_mincount': 94,
5416 'info_dict': {
5417 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5418 'title': 'Igor Kleiner - Playlists',
5419 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5420 'uploader': 'Igor Kleiner',
5421 'uploader_id': '@IgorDataScience',
5422 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5423 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5424 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5425 'channel': 'Igor Kleiner',
5426 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5427 'channel_follower_count': int
5428 },
5429 }, {
5430 'note': 'playlists, series',
5431 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5432 'playlist_mincount': 5,
5433 'info_dict': {
5434 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5435 'title': '3Blue1Brown - Playlists',
5436 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5437 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5438 'channel': '3Blue1Brown',
5439 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5440 'uploader_id': '@3blue1brown',
5441 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5442 'uploader': '3Blue1Brown',
5443 'tags': ['Mathematics'],
5444 'channel_follower_count': int,
5445 'channel_is_verified': True,
5446 },
5447 }, {
5448 'note': 'playlists, singlepage',
5449 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5450 'playlist_mincount': 4,
5451 'info_dict': {
5452 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5453 'title': 'ThirstForScience - Playlists',
5454 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5455 'uploader': 'ThirstForScience',
5456 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5457 'uploader_id': '@ThirstForScience',
5458 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5459 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5460 'tags': 'count:13',
5461 'channel': 'ThirstForScience',
5462 'channel_follower_count': int
5463 }
5464 }, {
5465 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5466 'only_matching': True,
5467 }, {
5468 'note': 'basic, single video playlist',
5469 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5470 'info_dict': {
5471 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5472 'title': 'youtube-dl public playlist',
5473 'description': '',
5474 'tags': [],
5475 'view_count': int,
5476 'modified_date': '20201130',
5477 'channel': 'Sergey M.',
5478 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5479 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5480 'availability': 'public',
5481 'uploader': 'Sergey M.',
5482 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5483 'uploader_id': '@sergeym.6173',
5484 },
5485 'playlist_count': 1,
5486 }, {
5487 'note': 'empty playlist',
5488 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5489 'info_dict': {
5490 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5491 'title': 'youtube-dl empty playlist',
5492 'tags': [],
5493 'channel': 'Sergey M.',
5494 'description': '',
5495 'modified_date': '20160902',
5496 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5497 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5498 'availability': 'public',
5499 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5500 'uploader_id': '@sergeym.6173',
5501 'uploader': 'Sergey M.',
5502 },
5503 'playlist_count': 0,
5504 }, {
5505 'note': 'Home tab',
5506 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5507 'info_dict': {
5508 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5509 'title': 'lex will - Home',
5510 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5511 'uploader': 'lex will',
5512 'uploader_id': '@lexwill718',
5513 'channel': 'lex will',
5514 'tags': ['bible', 'history', 'prophesy'],
5515 'uploader_url': 'https://www.youtube.com/@lexwill718',
5516 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5517 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5518 'channel_follower_count': int
5519 },
5520 'playlist_mincount': 2,
5521 }, {
5522 'note': 'Videos tab',
5523 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5524 'info_dict': {
5525 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5526 'title': 'lex will - Videos',
5527 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5528 'uploader': 'lex will',
5529 'uploader_id': '@lexwill718',
5530 'tags': ['bible', 'history', 'prophesy'],
5531 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5532 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5533 'uploader_url': 'https://www.youtube.com/@lexwill718',
5534 'channel': 'lex will',
5535 'channel_follower_count': int
5536 },
5537 'playlist_mincount': 975,
5538 }, {
5539 'note': 'Videos tab, sorted by popular',
5540 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5541 'info_dict': {
5542 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5543 'title': 'lex will - Videos',
5544 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5545 'uploader': 'lex will',
5546 'uploader_id': '@lexwill718',
5547 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5548 'uploader_url': 'https://www.youtube.com/@lexwill718',
5549 'channel': 'lex will',
5550 'tags': ['bible', 'history', 'prophesy'],
5551 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5552 'channel_follower_count': int
5553 },
5554 'playlist_mincount': 199,
5555 }, {
5556 'note': 'Playlists tab',
5557 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5558 'info_dict': {
5559 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5560 'title': 'lex will - Playlists',
5561 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5562 'uploader': 'lex will',
5563 'uploader_id': '@lexwill718',
5564 'uploader_url': 'https://www.youtube.com/@lexwill718',
5565 'channel': 'lex will',
5566 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5567 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5568 'tags': ['bible', 'history', 'prophesy'],
5569 'channel_follower_count': int
5570 },
5571 'playlist_mincount': 17,
5572 }, {
5573 'note': 'Community tab',
5574 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5575 'info_dict': {
5576 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5577 'title': 'lex will - Community',
5578 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5579 'channel': 'lex will',
5580 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5581 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5582 'tags': ['bible', 'history', 'prophesy'],
5583 'channel_follower_count': int,
5584 'uploader_url': 'https://www.youtube.com/@lexwill718',
5585 'uploader_id': '@lexwill718',
5586 'uploader': 'lex will',
5587 },
5588 'playlist_mincount': 18,
5589 }, {
5590 'note': 'Channels tab',
5591 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5592 'info_dict': {
5593 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5594 'title': 'lex will - Channels',
5595 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5596 'channel': 'lex will',
5597 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5598 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5599 'tags': ['bible', 'history', 'prophesy'],
5600 'channel_follower_count': int,
5601 'uploader_url': 'https://www.youtube.com/@lexwill718',
5602 'uploader_id': '@lexwill718',
5603 'uploader': 'lex will',
5604 },
5605 'playlist_mincount': 12,
5606 }, {
5607 'note': 'Search tab',
5608 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5609 'playlist_mincount': 40,
5610 'info_dict': {
5611 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5612 'title': '3Blue1Brown - Search - linear algebra',
5613 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5614 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5615 'tags': ['Mathematics'],
5616 'channel': '3Blue1Brown',
5617 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5618 'channel_follower_count': int,
5619 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5620 'uploader_id': '@3blue1brown',
5621 'uploader': '3Blue1Brown',
5622 'channel_is_verified': True,
5623 },
5624 }, {
5625 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5626 'only_matching': True,
5627 }, {
5628 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5629 'only_matching': True,
5630 }, {
5631 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5632 'only_matching': True,
5633 }, {
5634 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5635 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5636 'info_dict': {
5637 'title': '29C3: Not my department',
5638 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5639 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5640 'tags': [],
5641 'view_count': int,
5642 'modified_date': '20150605',
5643 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5644 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5645 'channel': 'Christiaan008',
5646 'availability': 'public',
5647 'uploader_id': '@ChRiStIaAn008',
5648 'uploader': 'Christiaan008',
5649 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5650 },
5651 'playlist_count': 96,
5652 }, {
5653 'note': 'Large playlist',
5654 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5655 'info_dict': {
5656 'title': 'Uploads from Cauchemar',
5657 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5658 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5659 'tags': [],
5660 'modified_date': r're:\d{8}',
5661 'channel': 'Cauchemar',
5662 'view_count': int,
5663 'description': '',
5664 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5665 'availability': 'public',
5666 'uploader_id': '@Cauchemar89',
5667 'uploader': 'Cauchemar',
5668 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5669 },
5670 'playlist_mincount': 1123,
5671 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5672 }, {
5673 'note': 'even larger playlist, 8832 videos',
5674 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5675 'only_matching': True,
5676 }, {
5677 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5678 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5679 'info_dict': {
5680 'title': 'Uploads from Interstellar Movie',
5681 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5682 'tags': [],
5683 'view_count': int,
5684 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5685 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5686 'channel': 'Interstellar Movie',
5687 'description': '',
5688 'modified_date': r're:\d{8}',
5689 'availability': 'public',
5690 'uploader_id': '@InterstellarMovie',
5691 'uploader': 'Interstellar Movie',
5692 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5693 },
5694 'playlist_mincount': 21,
5695 }, {
5696 'note': 'Playlist with "show unavailable videos" button',
5697 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5698 'info_dict': {
5699 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5700 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5701 'view_count': int,
5702 'channel': 'Phim Siêu Nhân Nhật Bản',
5703 'tags': [],
5704 'description': '',
5705 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5706 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5707 'modified_date': r're:\d{8}',
5708 'availability': 'public',
5709 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5710 'uploader_id': '@phimsieunhannhatban',
5711 'uploader': 'Phim Siêu Nhân Nhật Bản',
5712 },
5713 'playlist_mincount': 200,
5714 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5715 }, {
5716 'note': 'Playlist with unavailable videos in page 7',
5717 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5718 'info_dict': {
5719 'title': 'Uploads from BlankTV',
5720 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5721 'channel': 'BlankTV',
5722 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5723 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5724 'view_count': int,
5725 'tags': [],
5726 'modified_date': r're:\d{8}',
5727 'description': '',
5728 'availability': 'public',
5729 'uploader_id': '@blanktv',
5730 'uploader': 'BlankTV',
5731 'uploader_url': 'https://www.youtube.com/@blanktv',
5732 },
5733 'playlist_mincount': 1000,
5734 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5735 }, {
5736 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5737 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5738 'info_dict': {
5739 'title': 'Data Analysis with Dr Mike Pound',
5740 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5741 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5742 'tags': [],
5743 'view_count': int,
5744 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5745 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5746 'channel': 'Computerphile',
5747 'availability': 'public',
5748 'modified_date': '20190712',
5749 'uploader_id': '@Computerphile',
5750 'uploader': 'Computerphile',
5751 'uploader_url': 'https://www.youtube.com/@Computerphile',
5752 },
5753 'playlist_mincount': 11,
5754 }, {
5755 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5756 'only_matching': True,
5757 }, {
5758 'note': 'Playlist URL that does not actually serve a playlist',
5759 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5760 'info_dict': {
5761 'id': 'FqZTN594JQw',
5762 'ext': 'webm',
5763 'title': "Smiley's People 01 detective, Adventure Series, Action",
5764 'upload_date': '20150526',
5765 'license': 'Standard YouTube License',
5766 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5767 'categories': ['People & Blogs'],
5768 'tags': list,
5769 'view_count': int,
5770 'like_count': int,
5771 },
5772 'params': {
5773 'skip_download': True,
5774 },
5775 'skip': 'This video is not available.',
5776 'add_ie': [YoutubeIE.ie_key()],
5777 }, {
5778 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5779 'only_matching': True,
5780 }, {
5781 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5782 'only_matching': True,
5783 }, {
5784 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5785 'info_dict': {
5786 'id': 'hGkQjiJLjWQ', # This will keep changing
5787 'ext': 'mp4',
5788 'title': str,
5789 'upload_date': r're:\d{8}',
5790 'description': str,
5791 'categories': ['News & Politics'],
5792 'tags': list,
5793 'like_count': int,
5794 'release_timestamp': int,
5795 'channel': 'Sky News',
5796 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5797 'age_limit': 0,
5798 'view_count': int,
5799 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5800 'playable_in_embed': True,
5801 'release_date': r're:\d+',
5802 'availability': 'public',
5803 'live_status': 'is_live',
5804 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5805 'channel_follower_count': int,
5806 'concurrent_view_count': int,
5807 'uploader_url': 'https://www.youtube.com/@SkyNews',
5808 'uploader_id': '@SkyNews',
5809 'uploader': 'Sky News',
5810 'channel_is_verified': True,
5811 },
5812 'params': {
5813 'skip_download': True,
5814 },
5815 'expected_warnings': ['Ignoring subtitle tracks found in '],
5816 }, {
5817 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5818 'info_dict': {
5819 'id': 'a48o2S1cPoo',
5820 'ext': 'mp4',
5821 'title': 'The Young Turks - Live Main Show',
5822 'upload_date': '20150715',
5823 'license': 'Standard YouTube License',
5824 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5825 'categories': ['News & Politics'],
5826 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5827 'like_count': int,
5828 },
5829 'params': {
5830 'skip_download': True,
5831 },
5832 'only_matching': True,
5833 }, {
5834 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5835 'only_matching': True,
5836 }, {
5837 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5838 'only_matching': True,
5839 }, {
5840 'note': 'A channel that is not live. Should raise error',
5841 'url': 'https://www.youtube.com/user/numberphile/live',
5842 'only_matching': True,
5843 }, {
5844 'url': 'https://www.youtube.com/feed/trending',
5845 'only_matching': True,
5846 }, {
5847 'url': 'https://www.youtube.com/feed/library',
5848 'only_matching': True,
5849 }, {
5850 'url': 'https://www.youtube.com/feed/history',
5851 'only_matching': True,
5852 }, {
5853 'url': 'https://www.youtube.com/feed/subscriptions',
5854 'only_matching': True,
5855 }, {
5856 'url': 'https://www.youtube.com/feed/watch_later',
5857 'only_matching': True,
5858 }, {
5859 'note': 'Recommended - redirects to home page.',
5860 'url': 'https://www.youtube.com/feed/recommended',
5861 'only_matching': True,
5862 }, {
5863 'note': 'inline playlist with not always working continuations',
5864 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5865 'only_matching': True,
5866 }, {
5867 'url': 'https://www.youtube.com/course',
5868 'only_matching': True,
5869 }, {
5870 'url': 'https://www.youtube.com/zsecurity',
5871 'only_matching': True,
5872 }, {
5873 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5874 'only_matching': True,
5875 }, {
5876 'url': 'https://www.youtube.com/TheYoungTurks/live',
5877 'only_matching': True,
5878 }, {
5879 'url': 'https://www.youtube.com/hashtag/cctv9',
5880 'info_dict': {
5881 'id': 'cctv9',
5882 'title': '#cctv9',
5883 'tags': [],
5884 },
5885 'playlist_mincount': 300, # not consistent but should be over 300
5886 }, {
5887 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5888 'only_matching': True,
5889 }, {
5890 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5891 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5892 'only_matching': True
5893 }, {
5894 'note': '/browse/ should redirect to /channel/',
5895 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5896 'only_matching': True
5897 }, {
5898 'note': 'VLPL, should redirect to playlist?list=PL...',
5899 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5900 'info_dict': {
5901 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5902 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5903 'title': 'NCS : All Releases 💿',
5904 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5905 'modified_date': r're:\d{8}',
5906 'view_count': int,
5907 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5908 'tags': [],
5909 'channel': 'NoCopyrightSounds',
5910 'availability': 'public',
5911 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5912 'uploader': 'NoCopyrightSounds',
5913 'uploader_id': '@NoCopyrightSounds',
5914 },
5915 'playlist_mincount': 166,
5916 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5917 }, {
5918 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5919 'note': 'Topic, should redirect to playlist?list=UU...',
5920 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5921 'info_dict': {
5922 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5923 'title': 'Uploads from Royalty Free Music - Topic',
5924 'tags': [],
5925 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5926 'channel': 'Royalty Free Music - Topic',
5927 'view_count': int,
5928 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5929 'modified_date': r're:\d{8}',
5930 'description': '',
5931 'availability': 'public',
5932 'uploader': 'Royalty Free Music - Topic',
5933 },
5934 'playlist_mincount': 101,
5935 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5936 }, {
5937 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5938 # Treat as a general feed
5939 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5940 'info_dict': {
5941 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5942 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5943 'tags': [],
5944 },
5945 'playlist_mincount': 9,
5946 }, {
5947 'note': 'Youtube music Album',
5948 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5949 'info_dict': {
5950 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5951 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5952 'tags': [],
5953 'view_count': int,
5954 'description': '',
5955 'availability': 'unlisted',
5956 'modified_date': r're:\d{8}',
5957 },
5958 'playlist_count': 50,
5959 'expected_warnings': ['YouTube Music is not directly supported'],
5960 }, {
5961 'note': 'unlisted single video playlist',
5962 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5963 'info_dict': {
5964 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5965 'title': 'yt-dlp unlisted playlist test',
5966 'availability': 'unlisted',
5967 'tags': [],
5968 'modified_date': '20220418',
5969 'channel': 'colethedj',
5970 'view_count': int,
5971 'description': '',
5972 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5973 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5974 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5975 'uploader_id': '@colethedj1894',
5976 'uploader': 'colethedj',
5977 },
5978 'playlist': [{
5979 'info_dict': {
5980 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
5981 'id': 'BaW_jenozKc',
5982 '_type': 'url',
5983 'ie_key': 'Youtube',
5984 'duration': 10,
5985 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
5986 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
5987 'view_count': int,
5988 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
5989 'channel': 'Philipp Hagemeister',
5990 'uploader_id': '@PhilippHagemeister',
5991 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
5992 'uploader': 'Philipp Hagemeister',
5993 }
5994 }],
5995 'playlist_count': 1,
5996 'params': {'extract_flat': True},
5997 }, {
5998 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5999 'url': 'https://www.youtube.com/feed/recommended',
6000 'info_dict': {
6001 'id': 'recommended',
6002 'title': 'recommended',
6003 'tags': [],
6004 },
6005 'playlist_mincount': 50,
6006 'params': {
6007 'skip_download': True,
6008 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6009 },
6010 }, {
6011 'note': 'API Fallback: /videos tab, sorted by oldest first',
6012 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6013 'info_dict': {
6014 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6015 'title': 'Cody\'sLab - Videos',
6016 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6017 'channel': 'Cody\'sLab',
6018 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6019 'tags': [],
6020 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6021 'channel_follower_count': int
6022 },
6023 'playlist_mincount': 650,
6024 'params': {
6025 'skip_download': True,
6026 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6027 },
6028 'skip': 'Query for sorting no longer works',
6029 }, {
6030 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6031 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6032 'info_dict': {
6033 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6034 'title': 'Uploads from Royalty Free Music - Topic',
6035 'modified_date': r're:\d{8}',
6036 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6037 'description': '',
6038 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6039 'tags': [],
6040 'channel': 'Royalty Free Music - Topic',
6041 'view_count': int,
6042 'availability': 'public',
6043 'uploader': 'Royalty Free Music - Topic',
6044 },
6045 'playlist_mincount': 101,
6046 'params': {
6047 'skip_download': True,
6048 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6049 },
6050 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6051 }, {
6052 'note': 'non-standard redirect to regional channel',
6053 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6054 'only_matching': True
6055 }, {
6056 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6057 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6058 'info_dict': {
6059 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6060 'modified_date': '20220407',
6061 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6062 'tags': [],
6063 'availability': 'unlisted',
6064 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6065 'channel': 'pukkandan',
6066 'description': 'Test for collaborative playlist',
6067 'title': 'yt-dlp test - collaborative playlist',
6068 'view_count': int,
6069 'uploader_url': 'https://www.youtube.com/@pukkandan',
6070 'uploader_id': '@pukkandan',
6071 'uploader': 'pukkandan',
6072 },
6073 'playlist_mincount': 2
6074 }, {
6075 'note': 'translated tab name',
6076 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6077 'info_dict': {
6078 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6079 'tags': [],
6080 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6081 'description': 'test description',
6082 'title': 'cole-dlp-test-acc - 再生リスト',
6083 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6084 'channel': 'cole-dlp-test-acc',
6085 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6086 'uploader_id': '@coletdjnz',
6087 'uploader': 'cole-dlp-test-acc',
6088 },
6089 'playlist_mincount': 1,
6090 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6091 'expected_warnings': ['Preferring "ja"'],
6092 }, {
6093 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6094 'note': 'preferred lang set with playlist with translated video titles',
6095 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6096 'info_dict': {
6097 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6098 'tags': [],
6099 'view_count': int,
6100 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6101 'channel': 'cole-dlp-test-acc',
6102 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6103 'description': 'test',
6104 'title': 'dlp test playlist',
6105 'availability': 'public',
6106 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6107 'uploader_id': '@coletdjnz',
6108 'uploader': 'cole-dlp-test-acc',
6109 },
6110 'playlist_mincount': 1,
6111 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6112 'expected_warnings': ['Preferring "ja"'],
6113 }, {
6114 # shorts audio pivot for 2GtVksBMYFM.
6115 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6116 'info_dict': {
6117 'id': 'sfv_audio_pivot',
6118 'title': 'sfv_audio_pivot',
6119 'tags': [],
6120 },
6121 'playlist_mincount': 50,
6122
6123 }, {
6124 # Channel with a real live tab (not to be mistaken with streams tab)
6125 # Do not treat like it should redirect to live stream
6126 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6127 'info_dict': {
6128 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6129 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6130 'tags': [],
6131 },
6132 'playlist_mincount': 20,
6133 }, {
6134 # Tab name is not the same as tab id
6135 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6136 'info_dict': {
6137 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6138 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6139 'tags': [],
6140 },
6141 'playlist_mincount': 8,
6142 }, {
6143 # Home tab id is literally home. Not to get mistaken with featured
6144 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6145 'info_dict': {
6146 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6147 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6148 'tags': [],
6149 },
6150 'playlist_mincount': 8,
6151 }, {
6152 # Should get three playlists for videos, shorts and streams tabs
6153 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6154 'info_dict': {
6155 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6156 'title': 'Polka Ch. 尾丸ポルカ',
6157 'channel_follower_count': int,
6158 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6159 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6160 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6161 'channel': 'Polka Ch. 尾丸ポルカ',
6162 'tags': 'count:35',
6163 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6164 'uploader': 'Polka Ch. 尾丸ポルカ',
6165 'uploader_id': '@OmaruPolka',
6166 },
6167 'playlist_count': 3,
6168 }, {
6169 # Shorts tab with channel with handle
6170 # TODO: fix channel description
6171 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6172 'info_dict': {
6173 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6174 'title': 'Not Just Bikes - Shorts',
6175 'tags': 'count:12',
6176 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6177 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6178 'channel_follower_count': int,
6179 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6180 'channel': 'Not Just Bikes',
6181 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6182 'uploader': 'Not Just Bikes',
6183 'uploader_id': '@NotJustBikes',
6184 },
6185 'playlist_mincount': 10,
6186 }, {
6187 # Streams tab
6188 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6189 'info_dict': {
6190 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6191 'title': '中村悠一 - Live',
6192 'tags': 'count:7',
6193 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6194 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6195 'channel': '中村悠一',
6196 'channel_follower_count': int,
6197 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6198 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6199 'uploader_id': '@Yuichi-Nakamura',
6200 'uploader': '中村悠一',
6201 },
6202 'playlist_mincount': 60,
6203 }, {
6204 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6205 # See test_youtube_lists
6206 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6207 'only_matching': True,
6208 }, {
6209 # No uploads and no UCID given. Should fail with no uploads error
6210 # See test_youtube_lists
6211 'url': 'https://www.youtube.com/news',
6212 'only_matching': True
6213 }, {
6214 # No videos tab but has a shorts tab
6215 'url': 'https://www.youtube.com/c/TKFShorts',
6216 'info_dict': {
6217 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6218 'title': 'Shorts Break - Shorts',
6219 'tags': 'count:48',
6220 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6221 'channel': 'Shorts Break',
6222 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6223 'channel_follower_count': int,
6224 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6225 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6226 'uploader': 'Shorts Break',
6227 'uploader_id': '@ShortsBreak_Official',
6228 },
6229 'playlist_mincount': 30,
6230 }, {
6231 # Trending Now Tab. tab id is empty
6232 'url': 'https://www.youtube.com/feed/trending',
6233 'info_dict': {
6234 'id': 'trending',
6235 'title': 'trending - Now',
6236 'tags': [],
6237 },
6238 'playlist_mincount': 30,
6239 }, {
6240 # Trending Gaming Tab. tab id is empty
6241 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6242 'info_dict': {
6243 'id': 'trending',
6244 'title': 'trending - Gaming',
6245 'tags': [],
6246 },
6247 'playlist_mincount': 30,
6248 }, {
6249 # Shorts url result in shorts tab
6250 # TODO: Fix channel id extraction
6251 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6252 'info_dict': {
6253 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6254 'title': 'cole-dlp-test-acc - Shorts',
6255 'channel': 'cole-dlp-test-acc',
6256 'description': 'test description',
6257 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6258 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6259 'tags': [],
6260 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6261 'uploader_id': '@coletdjnz',
6262 'uploader': 'cole-dlp-test-acc',
6263 },
6264 'playlist': [{
6265 'info_dict': {
6266 # Channel data is not currently available for short renderers (as of 2023-03-01)
6267 '_type': 'url',
6268 'ie_key': 'Youtube',
6269 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6270 'id': 'sSM9J5YH_60',
6271 'title': 'SHORT short',
6272 'view_count': int,
6273 'thumbnails': list,
6274 }
6275 }],
6276 'params': {'extract_flat': True},
6277 }, {
6278 # Live video status should be extracted
6279 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6280 'info_dict': {
6281 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6282 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6283 'tags': []
6284 },
6285 'playlist': [{
6286 'info_dict': {
6287 '_type': 'url',
6288 'ie_key': 'Youtube',
6289 'url': 'startswith:https://www.youtube.com/watch?v=',
6290 'id': str,
6291 'title': str,
6292 'live_status': 'is_live',
6293 'channel_id': str,
6294 'channel_url': str,
6295 'concurrent_view_count': int,
6296 'channel': str,
6297 'uploader': str,
6298 'uploader_url': str,
6299 'uploader_id': str,
6300 'channel_is_verified': bool, # this will keep changing
6301 }
6302 }],
6303 'params': {'extract_flat': True, 'playlist_items': '1'},
6304 'playlist_mincount': 1
6305 }, {
6306 # Channel renderer metadata. Contains number of videos on the channel
6307 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6308 'info_dict': {
6309 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6310 'title': 'cole-dlp-test-acc - Channels',
6311 'channel': 'cole-dlp-test-acc',
6312 'description': 'test description',
6313 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6314 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6315 'tags': [],
6316 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6317 'uploader_id': '@coletdjnz',
6318 'uploader': 'cole-dlp-test-acc',
6319 },
6320 'playlist': [{
6321 'info_dict': {
6322 '_type': 'url',
6323 'ie_key': 'YoutubeTab',
6324 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6325 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6326 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6327 'title': 'PewDiePie',
6328 'channel': 'PewDiePie',
6329 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6330 'thumbnails': list,
6331 'channel_follower_count': int,
6332 'playlist_count': int,
6333 'uploader': 'PewDiePie',
6334 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6335 'uploader_id': '@PewDiePie',
6336 'channel_is_verified': True,
6337 }
6338 }],
6339 'params': {'extract_flat': True},
6340 }, {
6341 'url': 'https://www.youtube.com/@3blue1brown/about',
6342 'info_dict': {
6343 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6344 'tags': ['Mathematics'],
6345 'title': '3Blue1Brown - About',
6346 'channel_follower_count': int,
6347 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6348 'channel': '3Blue1Brown',
6349 'view_count': int,
6350 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6351 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6352 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6353 'uploader_id': '@3blue1brown',
6354 'uploader': '3Blue1Brown',
6355 'channel_is_verified': True,
6356 },
6357 'playlist_count': 0,
6358 }, {
6359 # Podcasts tab, with rich entry playlistRenderers
6360 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6361 'info_dict': {
6362 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6363 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6364 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6365 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6366 'title': '99 Percent Invisible - Podcasts',
6367 'uploader': '99 Percent Invisible',
6368 'channel_follower_count': int,
6369 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6370 'tags': [],
6371 'channel': '99 Percent Invisible',
6372 'uploader_id': '@99percentinvisiblepodcast',
6373 },
6374 'playlist_count': 1,
6375 }, {
6376 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6377 'url': 'https://www.youtube.com/@AHimitsu/releases',
6378 'info_dict': {
6379 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6380 'channel': 'A Himitsu',
6381 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6382 'title': 'A Himitsu - Releases',
6383 'uploader_id': '@AHimitsu',
6384 'uploader': 'A Himitsu',
6385 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6386 'tags': 'count:16',
6387 'description': 'I make music',
6388 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6389 'channel_follower_count': int,
6390 'channel_is_verified': True,
6391 },
6392 'playlist_mincount': 10,
6393 }]
6394
6395 @classmethod
6396 def suitable(cls, url):
6397 return False if YoutubeIE.suitable(url) else super().suitable(url)
6398
6399 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6400
6401 def _get_url_mobj(self, url):
6402 mobj = self._URL_RE.match(url).groupdict()
6403 mobj.update((k, '') for k, v in mobj.items() if v is None)
6404 return mobj
6405
6406 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6407 tab_name = (tab.get('title') or '').lower()
6408 tab_url = urljoin(base_url, traverse_obj(
6409 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6410
6411 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6412 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6413 if tab_id:
6414 return {
6415 'TAB_ID_SPONSORSHIPS': 'membership',
6416 }.get(tab_id, tab_id), tab_name
6417
6418 # Fallback to tab name if we cannot get the tab id.
6419 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6420 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6421 if tab_name:
6422 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6423 return {
6424 'home': 'featured',
6425 'live': 'streams',
6426 }.get(tab_name, tab_name), tab_name
6427
6428 def _has_tab(self, tabs, tab_id):
6429 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6430
6431 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6432 def _real_extract(self, url, smuggled_data):
6433 item_id = self._match_id(url)
6434 url = urllib.parse.urlunparse(
6435 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6436 compat_opts = self.get_param('compat_opts', [])
6437
6438 mobj = self._get_url_mobj(url)
6439 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6440 if is_channel and smuggled_data.get('is_music_url'):
6441 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6442 return self.url_result(
6443 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6444 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6445 mdata = self._extract_tab_endpoint(
6446 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6447 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6448 get_all=False, expected_type=str)
6449 if not murl:
6450 raise ExtractorError('Failed to resolve album to playlist')
6451 return self.url_result(murl, YoutubeTabIE)
6452 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6453 return self.url_result(
6454 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6455
6456 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6457 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6458 url = f'{pre}/videos{post}'
6459 if smuggled_data.get('is_music_url'):
6460 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6461
6462 # Handle both video/playlist URLs
6463 qs = parse_qs(url)
6464 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6465 if not video_id and mobj['not_channel'].startswith('watch'):
6466 if not playlist_id:
6467 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6468 raise ExtractorError('A video URL was given without video ID', expected=True)
6469 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6470 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6471 return self.url_result(
6472 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6473
6474 if not self._yes_playlist(playlist_id, video_id):
6475 return self.url_result(
6476 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6477
6478 data, ytcfg = self._extract_data(url, display_id)
6479
6480 # YouTube may provide a non-standard redirect to the regional channel
6481 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6482 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6483 redirect_url = traverse_obj(
6484 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6485 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6486 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6487 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6488 return self.url_result(redirect_url, YoutubeTabIE)
6489
6490 tabs, extra_tabs = self._extract_tab_renderers(data), []
6491 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6492 selected_tab = self._extract_selected_tab(tabs)
6493 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6494 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6495
6496 if not original_tab_id and selected_tab_name:
6497 self.to_screen('Downloading all uploads of the channel. '
6498 'To download only the videos in a specific tab, pass the tab\'s URL')
6499 if self._has_tab(tabs, 'streams'):
6500 extra_tabs.append(''.join((pre, '/streams', post)))
6501 if self._has_tab(tabs, 'shorts'):
6502 extra_tabs.append(''.join((pre, '/shorts', post)))
6503 # XXX: Members-only tab should also be extracted
6504
6505 if not extra_tabs and selected_tab_id != 'videos':
6506 # Channel does not have streams, shorts or videos tabs
6507 if item_id[:2] != 'UC':
6508 raise ExtractorError('This channel has no uploads', expected=True)
6509
6510 # Topic channels don't have /videos. Use the equivalent playlist instead
6511 pl_id = f'UU{item_id[2:]}'
6512 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6513 try:
6514 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6515 except ExtractorError:
6516 raise ExtractorError('This channel has no uploads', expected=True)
6517 else:
6518 item_id, url = pl_id, pl_url
6519 self.to_screen(
6520 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6521
6522 elif extra_tabs and selected_tab_id != 'videos':
6523 # When there are shorts/live tabs but not videos tab
6524 url, data = f'{pre}{post}', None
6525
6526 elif (original_tab_id or 'videos') != selected_tab_id:
6527 if original_tab_id == 'live':
6528 # Live tab should have redirected to the video
6529 # Except in the case the channel has an actual live tab
6530 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6531 raise UserNotLive(video_id=item_id)
6532 elif selected_tab_name:
6533 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6534
6535 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6536 url = f'{pre}{post}'
6537
6538 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6539 if 'no-youtube-unavailable-videos' not in compat_opts:
6540 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6541 self._extract_and_report_alerts(data, only_once=True)
6542
6543 tabs, entries = self._extract_tab_renderers(data), []
6544 if tabs:
6545 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6546 entries[0].update({
6547 'extractor_key': YoutubeTabIE.ie_key(),
6548 'extractor': YoutubeTabIE.IE_NAME,
6549 'webpage_url': url,
6550 })
6551 if self.get_param('playlist_items') == '0':
6552 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6553 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6554 entries.extend(map(self._real_extract, extra_tabs))
6555
6556 if len(entries) == 1:
6557 return entries[0]
6558 elif entries:
6559 metadata = self._extract_metadata_from_tabs(item_id, data)
6560 uploads_url = 'the Uploads (UU) playlist URL'
6561 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6562 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6563 self.to_screen(
6564 'Downloading as multiple playlists, separated by tabs. '
6565 f'To download as a single playlist instead, pass {uploads_url}')
6566 return self.playlist_result(entries, item_id, **metadata)
6567
6568 # Inline playlist
6569 playlist = traverse_obj(
6570 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6571 if playlist:
6572 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6573
6574 video_id = traverse_obj(
6575 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6576 if video_id:
6577 if tab != '/live': # live tab is expected to redirect to video
6578 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6579 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6580
6581 raise ExtractorError('Unable to recognize tab page')
6582
6583
6584 class YoutubePlaylistIE(InfoExtractor):
6585 IE_DESC = 'YouTube playlists'
6586 _VALID_URL = r'''(?x)(?:
6587 (?:https?://)?
6588 (?:\w+\.)?
6589 (?:
6590 (?:
6591 youtube(?:kids)?\.com|
6592 %(invidious)s
6593 )
6594 /.*?\?.*?\blist=
6595 )?
6596 (?P<id>%(playlist_id)s)
6597 )''' % {
6598 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6599 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6600 }
6601 IE_NAME = 'youtube:playlist'
6602 _TESTS = [{
6603 'note': 'issue #673',
6604 'url': 'PLBB231211A4F62143',
6605 'info_dict': {
6606 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6607 'id': 'PLBB231211A4F62143',
6608 'uploader': 'Wickman',
6609 'uploader_id': '@WickmanVT',
6610 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6611 'view_count': int,
6612 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6613 'modified_date': r're:\d{8}',
6614 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6615 'channel': 'Wickman',
6616 'tags': [],
6617 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6618 'availability': 'public',
6619 },
6620 'playlist_mincount': 29,
6621 }, {
6622 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6623 'info_dict': {
6624 'title': 'YDL_safe_search',
6625 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6626 },
6627 'playlist_count': 2,
6628 'skip': 'This playlist is private',
6629 }, {
6630 'note': 'embedded',
6631 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6632 'playlist_count': 4,
6633 'info_dict': {
6634 'title': 'JODA15',
6635 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6636 'uploader': 'milan',
6637 'uploader_id': '@milan5503',
6638 'description': '',
6639 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6640 'tags': [],
6641 'modified_date': '20140919',
6642 'view_count': int,
6643 'channel': 'milan',
6644 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6645 'uploader_url': 'https://www.youtube.com/@milan5503',
6646 'availability': 'public',
6647 },
6648 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6649 }, {
6650 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6651 'playlist_mincount': 455,
6652 'info_dict': {
6653 'title': '2018 Chinese New Singles (11/6 updated)',
6654 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6655 'uploader': 'LBK',
6656 'uploader_id': '@music_king',
6657 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6658 'channel': 'LBK',
6659 'view_count': int,
6660 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6661 'tags': [],
6662 'uploader_url': 'https://www.youtube.com/@music_king',
6663 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6664 'modified_date': r're:\d{8}',
6665 'availability': 'public',
6666 },
6667 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6668 }, {
6669 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6670 'only_matching': True,
6671 }, {
6672 # music album playlist
6673 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6674 'only_matching': True,
6675 }]
6676
6677 @classmethod
6678 def suitable(cls, url):
6679 if YoutubeTabIE.suitable(url):
6680 return False
6681 from ..utils import parse_qs
6682 qs = parse_qs(url)
6683 if qs.get('v', [None])[0]:
6684 return False
6685 return super().suitable(url)
6686
6687 def _real_extract(self, url):
6688 playlist_id = self._match_id(url)
6689 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6690 url = update_url_query(
6691 'https://www.youtube.com/playlist',
6692 parse_qs(url) or {'list': playlist_id})
6693 if is_music_url:
6694 url = smuggle_url(url, {'is_music_url': True})
6695 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6696
6697
6698 class YoutubeYtBeIE(InfoExtractor):
6699 IE_DESC = 'youtu.be'
6700 _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}
6701 _TESTS = [{
6702 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6703 'info_dict': {
6704 'id': 'yeWKywCrFtk',
6705 'ext': 'mp4',
6706 'title': 'Small Scale Baler and Braiding Rugs',
6707 'uploader': 'Backus-Page House Museum',
6708 'uploader_id': '@backuspagemuseum',
6709 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6710 'upload_date': '20161008',
6711 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6712 'categories': ['Nonprofits & Activism'],
6713 'tags': list,
6714 'like_count': int,
6715 'age_limit': 0,
6716 'playable_in_embed': True,
6717 'thumbnail': r're:^https?://.*\.webp',
6718 'channel': 'Backus-Page House Museum',
6719 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6720 'live_status': 'not_live',
6721 'view_count': int,
6722 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6723 'availability': 'public',
6724 'duration': 59,
6725 'comment_count': int,
6726 'channel_follower_count': int
6727 },
6728 'params': {
6729 'noplaylist': True,
6730 'skip_download': True,
6731 },
6732 }, {
6733 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6734 'only_matching': True,
6735 }]
6736
6737 def _real_extract(self, url):
6738 mobj = self._match_valid_url(url)
6739 video_id = mobj.group('id')
6740 playlist_id = mobj.group('playlist_id')
6741 return self.url_result(
6742 update_url_query('https://www.youtube.com/watch', {
6743 'v': video_id,
6744 'list': playlist_id,
6745 'feature': 'youtu.be',
6746 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6747
6748
6749 class YoutubeLivestreamEmbedIE(InfoExtractor):
6750 IE_DESC = 'YouTube livestream embeds'
6751 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6752 _TESTS = [{
6753 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6754 'only_matching': True,
6755 }]
6756
6757 def _real_extract(self, url):
6758 channel_id = self._match_id(url)
6759 return self.url_result(
6760 f'https://www.youtube.com/channel/{channel_id}/live',
6761 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6762
6763
6764 class YoutubeYtUserIE(InfoExtractor):
6765 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6766 IE_NAME = 'youtube:user'
6767 _VALID_URL = r'ytuser:(?P<id>.+)'
6768 _TESTS = [{
6769 'url': 'ytuser:phihag',
6770 'only_matching': True,
6771 }]
6772
6773 def _real_extract(self, url):
6774 user_id = self._match_id(url)
6775 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6776
6777
6778 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6779 IE_NAME = 'youtube:favorites'
6780 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6781 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6782 _LOGIN_REQUIRED = True
6783 _TESTS = [{
6784 'url': ':ytfav',
6785 'only_matching': True,
6786 }, {
6787 'url': ':ytfavorites',
6788 'only_matching': True,
6789 }]
6790
6791 def _real_extract(self, url):
6792 return self.url_result(
6793 'https://www.youtube.com/playlist?list=LL',
6794 ie=YoutubeTabIE.ie_key())
6795
6796
6797 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6798 IE_NAME = 'youtube:notif'
6799 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6800 _VALID_URL = r':ytnotif(?:ication)?s?'
6801 _LOGIN_REQUIRED = True
6802 _TESTS = [{
6803 'url': ':ytnotif',
6804 'only_matching': True,
6805 }, {
6806 'url': ':ytnotifications',
6807 'only_matching': True,
6808 }]
6809
6810 def _extract_notification_menu(self, response, continuation_list):
6811 notification_list = traverse_obj(
6812 response,
6813 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6814 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6815 expected_type=list) or []
6816 continuation_list[0] = None
6817 for item in notification_list:
6818 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6819 if entry:
6820 yield entry
6821 continuation = item.get('continuationItemRenderer')
6822 if continuation:
6823 continuation_list[0] = continuation
6824
6825 def _extract_notification_renderer(self, notification):
6826 video_id = traverse_obj(
6827 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6828 url = f'https://www.youtube.com/watch?v={video_id}'
6829 channel_id = None
6830 if not video_id:
6831 browse_ep = traverse_obj(
6832 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6833 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6834 post_id = self._search_regex(
6835 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6836 'post id', default=None)
6837 if not channel_id or not post_id:
6838 return
6839 # The direct /post url redirects to this in the browser
6840 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6841
6842 channel = traverse_obj(
6843 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6844 expected_type=str)
6845 notification_title = self._get_text(notification, 'shortMessage')
6846 if notification_title:
6847 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6848 # TODO: handle recommended videos
6849 title = self._search_regex(
6850 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6851 'video title', default=None)
6852 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6853 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6854 else None)
6855 return {
6856 '_type': 'url',
6857 'url': url,
6858 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6859 'video_id': video_id,
6860 'title': title,
6861 'channel_id': channel_id,
6862 'channel': channel,
6863 'uploader': channel,
6864 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6865 'timestamp': timestamp,
6866 }
6867
6868 def _notification_menu_entries(self, ytcfg):
6869 continuation_list = [None]
6870 response = None
6871 for page in itertools.count(1):
6872 ctoken = traverse_obj(
6873 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6874 response = self._extract_response(
6875 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6876 ep='notification/get_notification_menu', check_get_keys='actions',
6877 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6878 yield from self._extract_notification_menu(response, continuation_list)
6879 if not continuation_list[0]:
6880 break
6881
6882 def _real_extract(self, url):
6883 display_id = 'notifications'
6884 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6885 self._report_playlist_authcheck(ytcfg)
6886 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6887
6888
6889 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6890 IE_DESC = 'YouTube search'
6891 IE_NAME = 'youtube:search'
6892 _SEARCH_KEY = 'ytsearch'
6893 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6894 _TESTS = [{
6895 'url': 'ytsearch5:youtube-dl test video',
6896 'playlist_count': 5,
6897 'info_dict': {
6898 'id': 'youtube-dl test video',
6899 'title': 'youtube-dl test video',
6900 }
6901 }]
6902
6903
6904 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6905 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6906 _SEARCH_KEY = 'ytsearchdate'
6907 IE_DESC = 'YouTube search, newest videos first'
6908 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6909 _TESTS = [{
6910 'url': 'ytsearchdate5:youtube-dl test video',
6911 'playlist_count': 5,
6912 'info_dict': {
6913 'id': 'youtube-dl test video',
6914 'title': 'youtube-dl test video',
6915 }
6916 }]
6917
6918
6919 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6920 IE_DESC = 'YouTube search URLs with sorting and filter support'
6921 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6922 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6923 _TESTS = [{
6924 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6925 'playlist_mincount': 5,
6926 'info_dict': {
6927 'id': 'youtube-dl test video',
6928 'title': 'youtube-dl test video',
6929 }
6930 }, {
6931 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6932 'playlist_mincount': 5,
6933 'info_dict': {
6934 'id': 'python',
6935 'title': 'python',
6936 }
6937 }, {
6938 'url': 'https://www.youtube.com/results?search_query=%23cats',
6939 'playlist_mincount': 1,
6940 'info_dict': {
6941 'id': '#cats',
6942 'title': '#cats',
6943 # The test suite does not have support for nested playlists
6944 # 'entries': [{
6945 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6946 # 'title': '#cats',
6947 # }],
6948 },
6949 }, {
6950 # Channel results
6951 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6952 'info_dict': {
6953 'id': 'kurzgesagt',
6954 'title': 'kurzgesagt',
6955 },
6956 'playlist': [{
6957 'info_dict': {
6958 '_type': 'url',
6959 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6960 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6961 'ie_key': 'YoutubeTab',
6962 'channel': 'Kurzgesagt – In a Nutshell',
6963 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6964 'title': 'Kurzgesagt – In a Nutshell',
6965 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6966 # No longer available for search as it is set to the handle.
6967 # 'playlist_count': int,
6968 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6969 'thumbnails': list,
6970 'uploader_id': '@kurzgesagt',
6971 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
6972 'uploader': 'Kurzgesagt – In a Nutshell',
6973 'channel_is_verified': True,
6974 'channel_follower_count': int,
6975 }
6976 }],
6977 'params': {'extract_flat': True, 'playlist_items': '1'},
6978 'playlist_mincount': 1,
6979 }, {
6980 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
6981 'only_matching': True,
6982 }]
6983
6984 def _real_extract(self, url):
6985 qs = parse_qs(url)
6986 query = (qs.get('search_query') or qs.get('q'))[0]
6987 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
6988
6989
6990 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
6991 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
6992 IE_NAME = 'youtube:music:search_url'
6993 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6994 _TESTS = [{
6995 'url': 'https://music.youtube.com/search?q=royalty+free+music',
6996 'playlist_count': 16,
6997 'info_dict': {
6998 'id': 'royalty free music',
6999 'title': 'royalty free music',
7000 }
7001 }, {
7002 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7003 'playlist_mincount': 30,
7004 'info_dict': {
7005 'id': 'royalty free music - songs',
7006 'title': 'royalty free music - songs',
7007 },
7008 'params': {'extract_flat': 'in_playlist'}
7009 }, {
7010 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7011 'playlist_mincount': 30,
7012 'info_dict': {
7013 'id': 'royalty free music - community playlists',
7014 'title': 'royalty free music - community playlists',
7015 },
7016 'params': {'extract_flat': 'in_playlist'}
7017 }]
7018
7019 _SECTIONS = {
7020 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7021 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7022 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7023 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7024 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7025 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7026 }
7027
7028 def _real_extract(self, url):
7029 qs = parse_qs(url)
7030 query = (qs.get('search_query') or qs.get('q'))[0]
7031 params = qs.get('sp', (None,))[0]
7032 if params:
7033 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7034 else:
7035 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
7036 params = self._SECTIONS.get(section)
7037 if not params:
7038 section = None
7039 title = join_nonempty(query, section, delim=' - ')
7040 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7041
7042
7043 class YoutubeFeedsInfoExtractor(InfoExtractor):
7044 """
7045 Base class for feed extractors
7046 Subclasses must re-define the _FEED_NAME property.
7047 """
7048 _LOGIN_REQUIRED = True
7049 _FEED_NAME = 'feeds'
7050
7051 def _real_initialize(self):
7052 YoutubeBaseInfoExtractor._check_login_required(self)
7053
7054 @classproperty
7055 def IE_NAME(self):
7056 return f'youtube:{self._FEED_NAME}'
7057
7058 def _real_extract(self, url):
7059 return self.url_result(
7060 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7061
7062
7063 class YoutubeWatchLaterIE(InfoExtractor):
7064 IE_NAME = 'youtube:watchlater'
7065 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7066 _VALID_URL = r':ytwatchlater'
7067 _TESTS = [{
7068 'url': ':ytwatchlater',
7069 'only_matching': True,
7070 }]
7071
7072 def _real_extract(self, url):
7073 return self.url_result(
7074 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7075
7076
7077 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7078 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7079 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7080 _FEED_NAME = 'recommended'
7081 _LOGIN_REQUIRED = False
7082 _TESTS = [{
7083 'url': ':ytrec',
7084 'only_matching': True,
7085 }, {
7086 'url': ':ytrecommended',
7087 'only_matching': True,
7088 }, {
7089 'url': 'https://youtube.com',
7090 'only_matching': True,
7091 }]
7092
7093
7094 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7095 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7096 _VALID_URL = r':ytsub(?:scription)?s?'
7097 _FEED_NAME = 'subscriptions'
7098 _TESTS = [{
7099 'url': ':ytsubs',
7100 'only_matching': True,
7101 }, {
7102 'url': ':ytsubscriptions',
7103 'only_matching': True,
7104 }]
7105
7106
7107 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7108 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7109 _VALID_URL = r':ythis(?:tory)?'
7110 _FEED_NAME = 'history'
7111 _TESTS = [{
7112 'url': ':ythistory',
7113 'only_matching': True,
7114 }]
7115
7116
7117 class YoutubeStoriesIE(InfoExtractor):
7118 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
7119 IE_NAME = 'youtube:stories'
7120 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
7121 _TESTS = [{
7122 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
7123 'only_matching': True,
7124 }]
7125
7126 def _real_extract(self, url):
7127 playlist_id = f'RLTD{self._match_id(url)}'
7128 return self.url_result(
7129 smuggle_url(f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1', {'is_story': True}),
7130 ie=YoutubeTabIE, video_id=playlist_id)
7131
7132
7133 class YoutubeShortsAudioPivotIE(InfoExtractor):
7134 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7135 IE_NAME = 'youtube:shorts:pivot:audio'
7136 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7137 _TESTS = [{
7138 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7139 'only_matching': True,
7140 }]
7141
7142 @staticmethod
7143 def _generate_audio_pivot_params(video_id):
7144 """
7145 Generates sfv_audio_pivot browse params for this video id
7146 """
7147 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7148 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7149
7150 def _real_extract(self, url):
7151 video_id = self._match_id(url)
7152 return self.url_result(
7153 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7154 ie=YoutubeTabIE)
7155
7156
7157 class YoutubeTruncatedURLIE(InfoExtractor):
7158 IE_NAME = 'youtube:truncated_url'
7159 IE_DESC = False # Do not list
7160 _VALID_URL = r'''(?x)
7161 (?:https?://)?
7162 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7163 (?:watch\?(?:
7164 feature=[a-z_]+|
7165 annotation_id=annotation_[^&]+|
7166 x-yt-cl=[0-9]+|
7167 hl=[^&]*|
7168 t=[0-9]+
7169 )?
7170 |
7171 attribution_link\?a=[^&]+
7172 )
7173 $
7174 '''
7175
7176 _TESTS = [{
7177 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7178 'only_matching': True,
7179 }, {
7180 'url': 'https://www.youtube.com/watch?',
7181 'only_matching': True,
7182 }, {
7183 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7184 'only_matching': True,
7185 }, {
7186 'url': 'https://www.youtube.com/watch?feature=foo',
7187 'only_matching': True,
7188 }, {
7189 'url': 'https://www.youtube.com/watch?hl=en-GB',
7190 'only_matching': True,
7191 }, {
7192 'url': 'https://www.youtube.com/watch?t=2372',
7193 'only_matching': True,
7194 }]
7195
7196 def _real_extract(self, url):
7197 raise ExtractorError(
7198 'Did you forget to quote the URL? Remember that & is a meta '
7199 'character in most shells, so you want to put the URL in quotes, '
7200 'like youtube-dl '
7201 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7202 ' or simply youtube-dl BaW_jenozKc .',
7203 expected=True)
7204
7205
7206 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7207 IE_NAME = 'youtube:clip'
7208 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7209 _TESTS = [{
7210 # FIXME: Other metadata should be extracted from the clip, not from the base video
7211 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7212 'info_dict': {
7213 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7214 'ext': 'mp4',
7215 'section_start': 29.0,
7216 'section_end': 39.7,
7217 'duration': 10.7,
7218 'age_limit': 0,
7219 'availability': 'public',
7220 'categories': ['Gaming'],
7221 'channel': 'Scott The Woz',
7222 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7223 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7224 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7225 'like_count': int,
7226 'playable_in_embed': True,
7227 'tags': 'count:17',
7228 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7229 'title': 'Mobile Games on Console - Scott The Woz',
7230 'upload_date': '20210920',
7231 'uploader': 'Scott The Woz',
7232 'uploader_id': '@ScottTheWoz',
7233 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7234 'view_count': int,
7235 'live_status': 'not_live',
7236 'channel_follower_count': int,
7237 'chapters': 'count:20',
7238 'comment_count': int,
7239 'heatmap': 'count:100',
7240 }
7241 }]
7242
7243 def _real_extract(self, url):
7244 clip_id = self._match_id(url)
7245 _, data = self._extract_webpage(url, clip_id)
7246
7247 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7248 if not video_id:
7249 raise ExtractorError('Unable to find video ID')
7250
7251 clip_data = traverse_obj(data, (
7252 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7253 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7254 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7255 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7256
7257 return {
7258 '_type': 'url_transparent',
7259 'url': f'https://www.youtube.com/watch?v={video_id}',
7260 'ie_key': YoutubeIE.ie_key(),
7261 'id': clip_id,
7262 'section_start': int(clip_data['startTimeMs']) / 1000,
7263 'section_end': int(clip_data['endTimeMs']) / 1000,
7264 }
7265
7266
7267 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7268 IE_NAME = 'youtube:consent'
7269 IE_DESC = False # Do not list
7270 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7271 _TESTS = [{
7272 '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',
7273 'info_dict': {
7274 'id': 'qVv6vCqciTM',
7275 'ext': 'mp4',
7276 'age_limit': 0,
7277 'uploader_id': '@sana_natori',
7278 'comment_count': int,
7279 'chapters': 'count:13',
7280 'upload_date': '20221223',
7281 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7282 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7283 'uploader_url': 'https://www.youtube.com/@sana_natori',
7284 'like_count': int,
7285 'release_date': '20221223',
7286 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7287 '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】',
7288 'view_count': int,
7289 'playable_in_embed': True,
7290 'duration': 4438,
7291 'availability': 'public',
7292 'channel_follower_count': int,
7293 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7294 'categories': ['Entertainment'],
7295 'live_status': 'was_live',
7296 'release_timestamp': 1671793345,
7297 'channel': 'さなちゃんねる',
7298 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7299 'uploader': 'さなちゃんねる',
7300 'channel_is_verified': True,
7301 'heatmap': 'count:100',
7302 },
7303 'add_ie': ['Youtube'],
7304 'params': {'skip_download': 'Youtube'},
7305 }]
7306
7307 def _real_extract(self, url):
7308 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7309 if not redirect_url:
7310 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7311 return self.url_result(redirect_url)
7312
7313
7314 class YoutubeTruncatedIDIE(InfoExtractor):
7315 IE_NAME = 'youtube:truncated_id'
7316 IE_DESC = False # Do not list
7317 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7318
7319 _TESTS = [{
7320 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7321 'only_matching': True,
7322 }]
7323
7324 def _real_extract(self, url):
7325 video_id = self._match_id(url)
7326 raise ExtractorError(
7327 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7328 expected=True)