]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[cleanup] Minor fixes (See desc)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import copy
4 import datetime
5 import functools
6 import hashlib
7 import itertools
8 import json
9 import math
10 import os.path
11 import random
12 import re
13 import sys
14 import threading
15 import time
16 import traceback
17
18 from .common import InfoExtractor, SearchInfoExtractor
19 from ..compat import (
20 compat_chr,
21 compat_HTTPError,
22 compat_parse_qs,
23 compat_str,
24 compat_urllib_parse_unquote_plus,
25 compat_urllib_parse_urlencode,
26 compat_urllib_parse_urlparse,
27 compat_urlparse,
28 )
29 from ..jsinterp import JSInterpreter
30 from ..utils import (
31 NO_DEFAULT,
32 ExtractorError,
33 bug_reports_message,
34 clean_html,
35 datetime_from_str,
36 dict_get,
37 error_to_compat_str,
38 float_or_none,
39 format_field,
40 get_first,
41 int_or_none,
42 is_html,
43 join_nonempty,
44 js_to_json,
45 mimetype2ext,
46 network_exceptions,
47 orderedSet,
48 parse_codecs,
49 parse_count,
50 parse_duration,
51 parse_iso8601,
52 parse_qs,
53 qualities,
54 remove_end,
55 remove_start,
56 smuggle_url,
57 str_or_none,
58 str_to_int,
59 strftime_or_none,
60 traverse_obj,
61 try_get,
62 unescapeHTML,
63 unified_strdate,
64 unified_timestamp,
65 unsmuggle_url,
66 update_url_query,
67 url_or_none,
68 urljoin,
69 variadic,
70 )
71
72 # any clients starting with _ cannot be explicity requested by the user
73 INNERTUBE_CLIENTS = {
74 'web': {
75 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
76 'INNERTUBE_CONTEXT': {
77 'client': {
78 'clientName': 'WEB',
79 'clientVersion': '2.20211221.00.00',
80 }
81 },
82 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
83 },
84 'web_embedded': {
85 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
86 'INNERTUBE_CONTEXT': {
87 'client': {
88 'clientName': 'WEB_EMBEDDED_PLAYER',
89 'clientVersion': '1.20211215.00.01',
90 },
91 },
92 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
93 },
94 'web_music': {
95 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
96 'INNERTUBE_HOST': 'music.youtube.com',
97 'INNERTUBE_CONTEXT': {
98 'client': {
99 'clientName': 'WEB_REMIX',
100 'clientVersion': '1.20211213.00.00',
101 }
102 },
103 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
104 },
105 'web_creator': {
106 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
107 'INNERTUBE_CONTEXT': {
108 'client': {
109 'clientName': 'WEB_CREATOR',
110 'clientVersion': '1.20211220.02.00',
111 }
112 },
113 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
114 },
115 'android': {
116 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
117 'INNERTUBE_CONTEXT': {
118 'client': {
119 'clientName': 'ANDROID',
120 'clientVersion': '16.49',
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': '16.49',
132 },
133 },
134 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
135 'REQUIRE_JS_PLAYER': False
136 },
137 'android_music': {
138 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
139 'INNERTUBE_CONTEXT': {
140 'client': {
141 'clientName': 'ANDROID_MUSIC',
142 'clientVersion': '4.57',
143 }
144 },
145 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
146 'REQUIRE_JS_PLAYER': False
147 },
148 'android_creator': {
149 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
150 'INNERTUBE_CONTEXT': {
151 'client': {
152 'clientName': 'ANDROID_CREATOR',
153 'clientVersion': '21.47',
154 },
155 },
156 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
157 'REQUIRE_JS_PLAYER': False
158 },
159 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
160 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
161 'ios': {
162 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
163 'INNERTUBE_CONTEXT': {
164 'client': {
165 'clientName': 'IOS',
166 'clientVersion': '16.46',
167 'deviceModel': 'iPhone14,3',
168 }
169 },
170 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
171 'REQUIRE_JS_PLAYER': False
172 },
173 'ios_embedded': {
174 'INNERTUBE_CONTEXT': {
175 'client': {
176 'clientName': 'IOS_MESSAGES_EXTENSION',
177 'clientVersion': '16.46',
178 'deviceModel': 'iPhone14,3',
179 },
180 },
181 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
182 'REQUIRE_JS_PLAYER': False
183 },
184 'ios_music': {
185 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
186 'INNERTUBE_CONTEXT': {
187 'client': {
188 'clientName': 'IOS_MUSIC',
189 'clientVersion': '4.57',
190 },
191 },
192 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
193 'REQUIRE_JS_PLAYER': False
194 },
195 'ios_creator': {
196 'INNERTUBE_CONTEXT': {
197 'client': {
198 'clientName': 'IOS_CREATOR',
199 'clientVersion': '21.47',
200 },
201 },
202 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
203 'REQUIRE_JS_PLAYER': False
204 },
205 # mweb has 'ultralow' formats
206 # See: https://github.com/yt-dlp/yt-dlp/pull/557
207 'mweb': {
208 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
209 'INNERTUBE_CONTEXT': {
210 'client': {
211 'clientName': 'MWEB',
212 'clientVersion': '2.20211221.01.00',
213 }
214 },
215 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
216 },
217 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
218 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
219 'tv_embedded': {
220 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
221 'INNERTUBE_CONTEXT': {
222 'client': {
223 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
224 'clientVersion': '2.0',
225 },
226 },
227 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
228 },
229 }
230
231
232 def _split_innertube_client(client_name):
233 variant, *base = client_name.rsplit('.', 1)
234 if base:
235 return variant, base[0], variant
236 base, *variant = client_name.split('_', 1)
237 return client_name, base, variant[0] if variant else None
238
239
240 def build_innertube_clients():
241 THIRD_PARTY = {
242 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
243 }
244 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
245 priority = qualities(BASE_CLIENTS[::-1])
246
247 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
248 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
249 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
250 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
251 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
252
253 _, base_client, variant = _split_innertube_client(client)
254 ytcfg['priority'] = 10 * priority(base_client)
255
256 if not variant:
257 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
258 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
259 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
260 embedscreen['priority'] -= 3
261 elif variant == 'embedded':
262 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
263 ytcfg['priority'] -= 2
264 else:
265 ytcfg['priority'] -= 3
266
267
268 build_innertube_clients()
269
270
271 class YoutubeBaseInfoExtractor(InfoExtractor):
272 """Provide base functions for Youtube extractors"""
273
274 _RESERVED_NAMES = (
275 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
276 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
277 r'browse|oembed|get_video_info|iframe_api|s/player|'
278 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
279
280 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
281
282 # _NETRC_MACHINE = 'youtube'
283
284 # If True it will raise an error if no login info is provided
285 _LOGIN_REQUIRED = False
286
287 _INVIDIOUS_SITES = (
288 # invidious-redirect websites
289 r'(?:www\.)?redirect\.invidious\.io',
290 r'(?:(?:www|dev)\.)?invidio\.us',
291 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
292 r'(?:www\.)?invidious\.pussthecat\.org',
293 r'(?:www\.)?invidious\.zee\.li',
294 r'(?:www\.)?invidious\.ethibox\.fr',
295 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
296 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
297 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
298 # youtube-dl invidious instances list
299 r'(?:(?:www|no)\.)?invidiou\.sh',
300 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
301 r'(?:www\.)?invidious\.kabi\.tk',
302 r'(?:www\.)?invidious\.mastodon\.host',
303 r'(?:www\.)?invidious\.zapashcanon\.fr',
304 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
305 r'(?:www\.)?invidious\.tinfoil-hat\.net',
306 r'(?:www\.)?invidious\.himiko\.cloud',
307 r'(?:www\.)?invidious\.reallyancient\.tech',
308 r'(?:www\.)?invidious\.tube',
309 r'(?:www\.)?invidiou\.site',
310 r'(?:www\.)?invidious\.site',
311 r'(?:www\.)?invidious\.xyz',
312 r'(?:www\.)?invidious\.nixnet\.xyz',
313 r'(?:www\.)?invidious\.048596\.xyz',
314 r'(?:www\.)?invidious\.drycat\.fr',
315 r'(?:www\.)?inv\.skyn3t\.in',
316 r'(?:www\.)?tube\.poal\.co',
317 r'(?:www\.)?tube\.connect\.cafe',
318 r'(?:www\.)?vid\.wxzm\.sx',
319 r'(?:www\.)?vid\.mint\.lgbt',
320 r'(?:www\.)?vid\.puffyan\.us',
321 r'(?:www\.)?yewtu\.be',
322 r'(?:www\.)?yt\.elukerio\.org',
323 r'(?:www\.)?yt\.lelux\.fi',
324 r'(?:www\.)?invidious\.ggc-project\.de',
325 r'(?:www\.)?yt\.maisputain\.ovh',
326 r'(?:www\.)?ytprivate\.com',
327 r'(?:www\.)?invidious\.13ad\.de',
328 r'(?:www\.)?invidious\.toot\.koeln',
329 r'(?:www\.)?invidious\.fdn\.fr',
330 r'(?:www\.)?watch\.nettohikari\.com',
331 r'(?:www\.)?invidious\.namazso\.eu',
332 r'(?:www\.)?invidious\.silkky\.cloud',
333 r'(?:www\.)?invidious\.exonip\.de',
334 r'(?:www\.)?invidious\.riverside\.rocks',
335 r'(?:www\.)?invidious\.blamefran\.net',
336 r'(?:www\.)?invidious\.moomoo\.de',
337 r'(?:www\.)?ytb\.trom\.tf',
338 r'(?:www\.)?yt\.cyberhost\.uk',
339 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
340 r'(?:www\.)?qklhadlycap4cnod\.onion',
341 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
342 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
343 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
344 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
345 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
346 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
347 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
348 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
349 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
350 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
351 )
352
353 def _initialize_consent(self):
354 cookies = self._get_cookies('https://www.youtube.com/')
355 if cookies.get('__Secure-3PSID'):
356 return
357 consent_id = None
358 consent = cookies.get('CONSENT')
359 if consent:
360 if 'YES' in consent.value:
361 return
362 consent_id = self._search_regex(
363 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
364 if not consent_id:
365 consent_id = random.randint(100, 999)
366 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
367
368 def _initialize_pref(self):
369 cookies = self._get_cookies('https://www.youtube.com/')
370 pref_cookie = cookies.get('PREF')
371 pref = {}
372 if pref_cookie:
373 try:
374 pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
375 except ValueError:
376 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
377 pref.update({'hl': 'en', 'tz': 'UTC'})
378 self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
379
380 def _real_initialize(self):
381 self._initialize_pref()
382 self._initialize_consent()
383 self._check_login_required()
384
385 def _check_login_required(self):
386 if (self._LOGIN_REQUIRED
387 and self.get_param('cookiefile') is None
388 and self.get_param('cookiesfrombrowser') is None):
389 self.raise_login_required('Login details are needed to download this content', method='cookies')
390
391 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
392 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
393 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
394
395 def _get_default_ytcfg(self, client='web'):
396 return copy.deepcopy(INNERTUBE_CLIENTS[client])
397
398 def _get_innertube_host(self, client='web'):
399 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
400
401 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
402 # try_get but with fallback to default ytcfg client values when present
403 _func = lambda y: try_get(y, getter, expected_type)
404 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
405
406 def _extract_client_name(self, ytcfg, default_client='web'):
407 return self._ytcfg_get_safe(
408 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
409 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
410
411 def _extract_client_version(self, ytcfg, default_client='web'):
412 return self._ytcfg_get_safe(
413 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
414 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
415
416 def _extract_api_key(self, ytcfg=None, default_client='web'):
417 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
418
419 def _extract_context(self, ytcfg=None, default_client='web'):
420 context = get_first(
421 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
422 # Enforce language and tz for extraction
423 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
424 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
425 return context
426
427 _SAPISID = None
428
429 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
430 time_now = round(time.time())
431 if self._SAPISID is None:
432 yt_cookies = self._get_cookies('https://www.youtube.com')
433 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
434 # See: https://github.com/yt-dlp/yt-dlp/issues/393
435 sapisid_cookie = dict_get(
436 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
437 if sapisid_cookie and sapisid_cookie.value:
438 self._SAPISID = sapisid_cookie.value
439 self.write_debug('Extracted SAPISID cookie')
440 # SAPISID cookie is required if not already present
441 if not yt_cookies.get('SAPISID'):
442 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
443 self._set_cookie(
444 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
445 else:
446 self._SAPISID = False
447 if not self._SAPISID:
448 return None
449 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
450 sapisidhash = hashlib.sha1(
451 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
452 return f'SAPISIDHASH {time_now}_{sapisidhash}'
453
454 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
455 note='Downloading API JSON', errnote='Unable to download API page',
456 context=None, api_key=None, api_hostname=None, default_client='web'):
457
458 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
459 data.update(query)
460 real_headers = self.generate_api_headers(default_client=default_client)
461 real_headers.update({'content-type': 'application/json'})
462 if headers:
463 real_headers.update(headers)
464 return self._download_json(
465 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
466 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
467 data=json.dumps(data).encode('utf8'), headers=real_headers,
468 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
469
470 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
471 data = self._search_regex(
472 (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
473 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
474 if data:
475 return self._parse_json(data, item_id, fatal=fatal)
476
477 @staticmethod
478 def _extract_session_index(*data):
479 """
480 Index of current account in account list.
481 See: https://github.com/yt-dlp/yt-dlp/pull/519
482 """
483 for ytcfg in data:
484 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
485 if session_index is not None:
486 return session_index
487
488 # Deprecated?
489 def _extract_identity_token(self, ytcfg=None, webpage=None):
490 if ytcfg:
491 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
492 if token:
493 return token
494 if webpage:
495 return self._search_regex(
496 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
497 'identity token', default=None, fatal=False)
498
499 @staticmethod
500 def _extract_account_syncid(*args):
501 """
502 Extract syncId required to download private playlists of secondary channels
503 @params response and/or ytcfg
504 """
505 for data in args:
506 # ytcfg includes channel_syncid if on secondary channel
507 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
508 if delegated_sid:
509 return delegated_sid
510 sync_ids = (try_get(
511 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
512 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
513 if len(sync_ids) >= 2 and sync_ids[1]:
514 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
515 # and just "user_syncid||" for primary channel. We only want the channel_syncid
516 return sync_ids[0]
517
518 @staticmethod
519 def _extract_visitor_data(*args):
520 """
521 Extracts visitorData from an API response or ytcfg
522 Appears to be used to track session state
523 """
524 return get_first(
525 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
526 expected_type=str)
527
528 @property
529 def is_authenticated(self):
530 return bool(self._generate_sapisidhash_header())
531
532 def extract_ytcfg(self, video_id, webpage):
533 if not webpage:
534 return {}
535 return self._parse_json(
536 self._search_regex(
537 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
538 default='{}'), video_id, fatal=False) or {}
539
540 def generate_api_headers(
541 self, *, ytcfg=None, account_syncid=None, session_index=None,
542 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
543
544 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
545 headers = {
546 'X-YouTube-Client-Name': compat_str(
547 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
548 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
549 'Origin': origin,
550 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
551 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
552 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
553 }
554 if session_index is None:
555 session_index = self._extract_session_index(ytcfg)
556 if account_syncid or session_index is not None:
557 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
558
559 auth = self._generate_sapisidhash_header(origin)
560 if auth is not None:
561 headers['Authorization'] = auth
562 headers['X-Origin'] = origin
563 return {h: v for h, v in headers.items() if v is not None}
564
565 def _download_ytcfg(self, client, video_id):
566 url = {
567 'web': 'https://www.youtube.com',
568 'web_music': 'https://music.youtube.com',
569 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
570 }.get(client)
571 if not url:
572 return {}
573 webpage = self._download_webpage(
574 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
575 return self.extract_ytcfg(video_id, webpage) or {}
576
577 @staticmethod
578 def _build_api_continuation_query(continuation, ctp=None):
579 query = {
580 'continuation': continuation
581 }
582 # TODO: Inconsistency with clickTrackingParams.
583 # Currently we have a fixed ctp contained within context (from ytcfg)
584 # and a ctp in root query for continuation.
585 if ctp:
586 query['clickTracking'] = {'clickTrackingParams': ctp}
587 return query
588
589 @classmethod
590 def _extract_next_continuation_data(cls, renderer):
591 next_continuation = try_get(
592 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
593 lambda x: x['continuation']['reloadContinuationData']), dict)
594 if not next_continuation:
595 return
596 continuation = next_continuation.get('continuation')
597 if not continuation:
598 return
599 ctp = next_continuation.get('clickTrackingParams')
600 return cls._build_api_continuation_query(continuation, ctp)
601
602 @classmethod
603 def _extract_continuation_ep_data(cls, continuation_ep: dict):
604 if isinstance(continuation_ep, dict):
605 continuation = try_get(
606 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
607 if not continuation:
608 return
609 ctp = continuation_ep.get('clickTrackingParams')
610 return cls._build_api_continuation_query(continuation, ctp)
611
612 @classmethod
613 def _extract_continuation(cls, renderer):
614 next_continuation = cls._extract_next_continuation_data(renderer)
615 if next_continuation:
616 return next_continuation
617
618 contents = []
619 for key in ('contents', 'items'):
620 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
621
622 for content in contents:
623 if not isinstance(content, dict):
624 continue
625 continuation_ep = try_get(
626 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
627 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
628 dict)
629 continuation = cls._extract_continuation_ep_data(continuation_ep)
630 if continuation:
631 return continuation
632
633 @classmethod
634 def _extract_alerts(cls, data):
635 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
636 if not isinstance(alert_dict, dict):
637 continue
638 for alert in alert_dict.values():
639 alert_type = alert.get('type')
640 if not alert_type:
641 continue
642 message = cls._get_text(alert, 'text')
643 if message:
644 yield alert_type, message
645
646 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
647 errors = []
648 warnings = []
649 for alert_type, alert_message in alerts:
650 if alert_type.lower() == 'error' and fatal:
651 errors.append([alert_type, alert_message])
652 else:
653 warnings.append([alert_type, alert_message])
654
655 for alert_type, alert_message in (warnings + errors[:-1]):
656 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
657 if errors:
658 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
659
660 def _extract_and_report_alerts(self, data, *args, **kwargs):
661 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
662
663 def _extract_badges(self, renderer: dict):
664 badges = set()
665 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
666 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
667 if label:
668 badges.add(label.lower())
669 return badges
670
671 @staticmethod
672 def _get_text(data, *path_list, max_runs=None):
673 for path in path_list or [None]:
674 if path is None:
675 obj = [data]
676 else:
677 obj = traverse_obj(data, path, default=[])
678 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
679 obj = [obj]
680 for item in obj:
681 text = try_get(item, lambda x: x['simpleText'], compat_str)
682 if text:
683 return text
684 runs = try_get(item, lambda x: x['runs'], list) or []
685 if not runs and isinstance(item, list):
686 runs = item
687
688 runs = runs[:min(len(runs), max_runs or len(runs))]
689 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
690 if text:
691 return text
692
693 def _get_count(self, data, *path_list):
694 count_text = self._get_text(data, *path_list) or ''
695 count = parse_count(count_text)
696 if count is None:
697 count = str_to_int(
698 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
699 return count
700
701 @staticmethod
702 def _extract_thumbnails(data, *path_list):
703 """
704 Extract thumbnails from thumbnails dict
705 @param path_list: path list to level that contains 'thumbnails' key
706 """
707 thumbnails = []
708 for path in path_list or [()]:
709 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
710 thumbnail_url = url_or_none(thumbnail.get('url'))
711 if not thumbnail_url:
712 continue
713 # Sometimes youtube gives a wrong thumbnail URL. See:
714 # https://github.com/yt-dlp/yt-dlp/issues/233
715 # https://github.com/ytdl-org/youtube-dl/issues/28023
716 if 'maxresdefault' in thumbnail_url:
717 thumbnail_url = thumbnail_url.split('?')[0]
718 thumbnails.append({
719 'url': thumbnail_url,
720 'height': int_or_none(thumbnail.get('height')),
721 'width': int_or_none(thumbnail.get('width')),
722 })
723 return thumbnails
724
725 @staticmethod
726 def extract_relative_time(relative_time_text):
727 """
728 Extracts a relative time from string and converts to dt object
729 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
730 """
731 mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
732 if mobj:
733 start = mobj.group('start')
734 if start:
735 return datetime_from_str(start)
736 try:
737 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
738 except ValueError:
739 return None
740
741 def _extract_time_text(self, renderer, *path_list):
742 """@returns (timestamp, time_text)"""
743 text = self._get_text(renderer, *path_list) or ''
744 dt = self.extract_relative_time(text)
745 timestamp = None
746 if isinstance(dt, datetime.datetime):
747 timestamp = calendar.timegm(dt.timetuple())
748
749 if timestamp is None:
750 timestamp = (
751 unified_timestamp(text) or unified_timestamp(
752 self._search_regex(
753 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
754 text.lower(), 'time text', default=None)))
755
756 if text and timestamp is None:
757 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
758 return timestamp, text
759
760 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
761 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
762 default_client='web'):
763 response = None
764 last_error = None
765 count = -1
766 retries = self.get_param('extractor_retries', 3)
767 if check_get_keys is None:
768 check_get_keys = []
769 while count < retries:
770 count += 1
771 if last_error:
772 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
773 try:
774 response = self._call_api(
775 ep=ep, fatal=True, headers=headers,
776 video_id=item_id, query=query,
777 context=self._extract_context(ytcfg, default_client),
778 api_key=self._extract_api_key(ytcfg, default_client),
779 api_hostname=api_hostname, default_client=default_client,
780 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
781 except ExtractorError as e:
782 if isinstance(e.cause, network_exceptions):
783 if isinstance(e.cause, compat_HTTPError):
784 first_bytes = e.cause.read(512)
785 if not is_html(first_bytes):
786 yt_error = try_get(
787 self._parse_json(
788 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
789 lambda x: x['error']['message'], compat_str)
790 if yt_error:
791 self._report_alerts([('ERROR', yt_error)], fatal=False)
792 # Downloading page may result in intermittent 5xx HTTP error
793 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
794 # We also want to catch all other network exceptions since errors in later pages can be troublesome
795 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
796 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
797 last_error = error_to_compat_str(e.cause or e.msg)
798 if count < retries:
799 continue
800 if fatal:
801 raise
802 else:
803 self.report_warning(error_to_compat_str(e))
804 return
805
806 else:
807 try:
808 self._extract_and_report_alerts(response, only_once=True)
809 except ExtractorError as e:
810 # YouTube servers may return errors we want to retry on in a 200 OK response
811 # See: https://github.com/yt-dlp/yt-dlp/issues/839
812 if 'unknown error' in e.msg.lower():
813 last_error = e.msg
814 continue
815 if fatal:
816 raise
817 self.report_warning(error_to_compat_str(e))
818 return
819 if not check_get_keys or dict_get(response, check_get_keys):
820 break
821 # Youtube sometimes sends incomplete data
822 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
823 last_error = 'Incomplete data received'
824 if count >= retries:
825 if fatal:
826 raise ExtractorError(last_error)
827 else:
828 self.report_warning(last_error)
829 return
830 return response
831
832 @staticmethod
833 def is_music_url(url):
834 return re.match(r'https?://music\.youtube\.com/', url) is not None
835
836 def _extract_video(self, renderer):
837 video_id = renderer.get('videoId')
838 title = self._get_text(renderer, 'title')
839 description = self._get_text(renderer, 'descriptionSnippet')
840 duration = parse_duration(self._get_text(
841 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
842 if duration is None:
843 duration = parse_duration(self._search_regex(
844 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
845 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
846 video_id, default=None, group='duration'))
847
848 view_count = self._get_count(renderer, 'viewCountText')
849
850 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
851 channel_id = traverse_obj(
852 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
853 expected_type=str, get_all=False)
854 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
855 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
856 overlay_style = traverse_obj(
857 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
858 get_all=False, expected_type=str)
859 badges = self._extract_badges(renderer)
860 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
861 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
862 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
863 expected_type=str)) or ''
864 url = f'https://www.youtube.com/watch?v={video_id}'
865 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
866 url = f'https://www.youtube.com/shorts/{video_id}'
867
868 return {
869 '_type': 'url',
870 'ie_key': YoutubeIE.ie_key(),
871 'id': video_id,
872 'url': url,
873 'title': title,
874 'description': description,
875 'duration': duration,
876 'view_count': view_count,
877 'uploader': uploader,
878 'channel_id': channel_id,
879 'thumbnails': thumbnails,
880 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
881 if self._configuration_arg('approximate_date', ie_key='youtubetab')
882 else None),
883 'live_status': ('is_upcoming' if scheduled_timestamp is not None
884 else 'was_live' if 'streamed' in time_text.lower()
885 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
886 else None),
887 'release_timestamp': scheduled_timestamp,
888 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
889 }
890
891
892 class YoutubeIE(YoutubeBaseInfoExtractor):
893 IE_DESC = 'YouTube'
894 _VALID_URL = r"""(?x)^
895 (
896 (?:https?://|//) # http(s):// or protocol-independent URL
897 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
898 (?:www\.)?deturl\.com/www\.youtube\.com|
899 (?:www\.)?pwnyoutube\.com|
900 (?:www\.)?hooktube\.com|
901 (?:www\.)?yourepeat\.com|
902 tube\.majestyc\.net|
903 %(invidious)s|
904 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
905 (?:.*?\#/)? # handle anchor (#/) redirect urls
906 (?: # the various things that can precede the ID:
907 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
908 |(?: # or the v= param in all its forms
909 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
910 (?:\?|\#!?) # the params delimiter ? or # or #!
911 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
912 v=
913 )
914 ))
915 |(?:
916 youtu\.be| # just youtu.be/xxxx
917 vid\.plus| # or vid.plus/xxxx
918 zwearz\.com/watch| # or zwearz.com/watch/xxxx
919 %(invidious)s
920 )/
921 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
922 )
923 )? # all until now is optional -> you can pass the naked ID
924 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
925 (?(1).+)? # if we found the ID, everything can follow
926 (?:\#|$)""" % {
927 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
928 }
929 _PLAYER_INFO_RE = (
930 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
931 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$',
932 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
933 )
934 _formats = {
935 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
936 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
937 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
938 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
939 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
940 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
941 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
942 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
943 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
944 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
945 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
946 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
947 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
948 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
949 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
950 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
951 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
952 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
953
954
955 # 3D videos
956 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
957 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
958 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
959 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
960 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
961 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
962 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
963
964 # Apple HTTP Live Streaming
965 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
966 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
967 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
968 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
969 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
970 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
971 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
972 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
973
974 # DASH mp4 video
975 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
976 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
977 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
978 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
979 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
980 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
981 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
982 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
983 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
984 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
985 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
986 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
987
988 # Dash mp4 audio
989 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
990 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
991 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
992 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
993 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
994 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
995 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
996
997 # Dash webm
998 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
999 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1000 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1001 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1002 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1003 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1004 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1005 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1006 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1007 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1008 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1009 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1010 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1011 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1012 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1013 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1014 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1015 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1016 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1017 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1018 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1019 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1020
1021 # Dash webm audio
1022 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1023 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1024
1025 # Dash webm audio with opus inside
1026 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1027 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1028 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1029
1030 # RTMP (unnamed)
1031 '_rtmp': {'protocol': 'rtmp'},
1032
1033 # av01 video only formats sometimes served with "unknown" codecs
1034 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1035 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1036 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1037 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1038 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1039 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1040 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1041 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1042 }
1043 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1044
1045 _GEO_BYPASS = False
1046
1047 IE_NAME = 'youtube'
1048 _TESTS = [
1049 {
1050 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1051 'info_dict': {
1052 'id': 'BaW_jenozKc',
1053 'ext': 'mp4',
1054 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1055 'uploader': 'Philipp Hagemeister',
1056 'uploader_id': 'phihag',
1057 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1058 'channel': 'Philipp Hagemeister',
1059 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1060 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1061 'upload_date': '20121002',
1062 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1063 'categories': ['Science & Technology'],
1064 'tags': ['youtube-dl'],
1065 'duration': 10,
1066 'view_count': int,
1067 'like_count': int,
1068 'availability': 'public',
1069 'playable_in_embed': True,
1070 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1071 'live_status': 'not_live',
1072 'age_limit': 0,
1073 'start_time': 1,
1074 'end_time': 9,
1075 'channel_follower_count': int
1076 }
1077 },
1078 {
1079 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1080 'note': 'Embed-only video (#1746)',
1081 'info_dict': {
1082 'id': 'yZIXLfi8CZQ',
1083 'ext': 'mp4',
1084 'upload_date': '20120608',
1085 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1086 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1087 'uploader': 'SET India',
1088 'uploader_id': 'setindia',
1089 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1090 'age_limit': 18,
1091 },
1092 'skip': 'Private video',
1093 },
1094 {
1095 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1096 'note': 'Use the first video ID in the URL',
1097 'info_dict': {
1098 'id': 'BaW_jenozKc',
1099 'ext': 'mp4',
1100 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1101 'uploader': 'Philipp Hagemeister',
1102 'uploader_id': 'phihag',
1103 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1104 'channel': 'Philipp Hagemeister',
1105 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1106 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1107 'upload_date': '20121002',
1108 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1109 'categories': ['Science & Technology'],
1110 'tags': ['youtube-dl'],
1111 'duration': 10,
1112 'view_count': int,
1113 'like_count': int,
1114 'availability': 'public',
1115 'playable_in_embed': True,
1116 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1117 'live_status': 'not_live',
1118 'age_limit': 0,
1119 'channel_follower_count': int
1120 },
1121 'params': {
1122 'skip_download': True,
1123 },
1124 },
1125 {
1126 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1127 'note': '256k DASH audio (format 141) via DASH manifest',
1128 'info_dict': {
1129 'id': 'a9LDPn-MO4I',
1130 'ext': 'm4a',
1131 'upload_date': '20121002',
1132 'uploader_id': '8KVIDEO',
1133 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1134 'description': '',
1135 'uploader': '8KVIDEO',
1136 'title': 'UHDTV TEST 8K VIDEO.mp4'
1137 },
1138 'params': {
1139 'youtube_include_dash_manifest': True,
1140 'format': '141',
1141 },
1142 'skip': 'format 141 not served anymore',
1143 },
1144 # DASH manifest with encrypted signature
1145 {
1146 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1147 'info_dict': {
1148 'id': 'IB3lcPjvWLA',
1149 'ext': 'm4a',
1150 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1151 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1152 'duration': 244,
1153 'uploader': 'AfrojackVEVO',
1154 'uploader_id': 'AfrojackVEVO',
1155 'upload_date': '20131011',
1156 'abr': 129.495,
1157 'like_count': int,
1158 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1159 'playable_in_embed': True,
1160 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1161 'view_count': int,
1162 'track': 'The Spark',
1163 'live_status': 'not_live',
1164 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1165 'channel': 'Afrojack',
1166 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1167 'tags': 'count:19',
1168 'availability': 'public',
1169 'categories': ['Music'],
1170 'age_limit': 0,
1171 'alt_title': 'The Spark',
1172 'channel_follower_count': int
1173 },
1174 'params': {
1175 'youtube_include_dash_manifest': True,
1176 'format': '141/bestaudio[ext=m4a]',
1177 },
1178 },
1179 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1180 {
1181 'note': 'Embed allowed age-gate video',
1182 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1183 'info_dict': {
1184 'id': 'HtVdAasjOgU',
1185 'ext': 'mp4',
1186 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1187 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1188 'duration': 142,
1189 'uploader': 'The Witcher',
1190 'uploader_id': 'WitcherGame',
1191 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1192 'upload_date': '20140605',
1193 'age_limit': 18,
1194 'categories': ['Gaming'],
1195 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1196 'availability': 'needs_auth',
1197 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1198 'like_count': int,
1199 'channel': 'The Witcher',
1200 'live_status': 'not_live',
1201 'tags': 'count:17',
1202 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1203 'playable_in_embed': True,
1204 'view_count': int,
1205 'channel_follower_count': int
1206 },
1207 },
1208 {
1209 'note': 'Age-gate video with embed allowed in public site',
1210 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1211 'info_dict': {
1212 'id': 'HsUATh_Nc2U',
1213 'ext': 'mp4',
1214 'title': 'Godzilla 2 (Official Video)',
1215 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1216 'upload_date': '20200408',
1217 'uploader_id': 'FlyingKitty900',
1218 'uploader': 'FlyingKitty',
1219 'age_limit': 18,
1220 'availability': 'needs_auth',
1221 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1222 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1223 'channel': 'FlyingKitty',
1224 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1225 'view_count': int,
1226 'categories': ['Entertainment'],
1227 'live_status': 'not_live',
1228 'tags': ['Flyingkitty', 'godzilla 2'],
1229 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1230 'like_count': int,
1231 'duration': 177,
1232 'playable_in_embed': True,
1233 'channel_follower_count': int
1234 },
1235 },
1236 {
1237 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1238 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1239 'info_dict': {
1240 'id': 'Tq92D6wQ1mg',
1241 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1242 'ext': 'mp4',
1243 'upload_date': '20191228',
1244 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1245 'uploader': 'Projekt Melody',
1246 'description': 'md5:17eccca93a786d51bc67646756894066',
1247 'age_limit': 18,
1248 'like_count': int,
1249 'availability': 'needs_auth',
1250 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1251 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1252 'view_count': int,
1253 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1254 'channel': 'Projekt Melody',
1255 'live_status': 'not_live',
1256 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1257 'playable_in_embed': True,
1258 'categories': ['Entertainment'],
1259 'duration': 106,
1260 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1261 'channel_follower_count': int
1262 },
1263 },
1264 {
1265 'note': 'Non-Agegated non-embeddable video',
1266 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1267 'info_dict': {
1268 'id': 'MeJVWBSsPAY',
1269 'ext': 'mp4',
1270 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1271 'uploader': 'Herr Lurik',
1272 'uploader_id': 'st3in234',
1273 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1274 'upload_date': '20130730',
1275 'track': 'Such mich find mich',
1276 'age_limit': 0,
1277 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1278 'like_count': int,
1279 'playable_in_embed': False,
1280 'creator': 'OOMPH!',
1281 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1282 'view_count': int,
1283 'alt_title': 'Such mich find mich',
1284 'duration': 210,
1285 'channel': 'Herr Lurik',
1286 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1287 'categories': ['Music'],
1288 'availability': 'public',
1289 'uploader_url': 'http://www.youtube.com/user/st3in234',
1290 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1291 'live_status': 'not_live',
1292 'artist': 'OOMPH!',
1293 'channel_follower_count': int
1294 },
1295 },
1296 {
1297 'note': 'Non-bypassable age-gated video',
1298 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1299 'only_matching': True,
1300 },
1301 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1302 # YouTube Red ad is not captured for creator
1303 {
1304 'url': '__2ABJjxzNo',
1305 'info_dict': {
1306 'id': '__2ABJjxzNo',
1307 'ext': 'mp4',
1308 'duration': 266,
1309 'upload_date': '20100430',
1310 'uploader_id': 'deadmau5',
1311 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1312 'creator': 'deadmau5',
1313 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1314 'uploader': 'deadmau5',
1315 'title': 'Deadmau5 - Some Chords (HD)',
1316 'alt_title': 'Some Chords',
1317 'availability': 'public',
1318 'tags': 'count:14',
1319 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1320 'view_count': int,
1321 'live_status': 'not_live',
1322 'channel': 'deadmau5',
1323 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1324 'like_count': int,
1325 'track': 'Some Chords',
1326 'artist': 'deadmau5',
1327 'playable_in_embed': True,
1328 'age_limit': 0,
1329 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1330 'categories': ['Music'],
1331 'album': 'Some Chords',
1332 'channel_follower_count': int
1333 },
1334 'expected_warnings': [
1335 'DASH manifest missing',
1336 ]
1337 },
1338 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1339 {
1340 'url': 'lqQg6PlCWgI',
1341 'info_dict': {
1342 'id': 'lqQg6PlCWgI',
1343 'ext': 'mp4',
1344 'duration': 6085,
1345 'upload_date': '20150827',
1346 'uploader_id': 'olympic',
1347 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1348 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1349 'uploader': 'Olympics',
1350 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1351 'like_count': int,
1352 'release_timestamp': 1343767800,
1353 'playable_in_embed': True,
1354 'categories': ['Sports'],
1355 'release_date': '20120731',
1356 'channel': 'Olympics',
1357 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1358 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1359 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1360 'age_limit': 0,
1361 'availability': 'public',
1362 'live_status': 'was_live',
1363 'view_count': int,
1364 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1365 'channel_follower_count': int
1366 },
1367 'params': {
1368 'skip_download': 'requires avconv',
1369 }
1370 },
1371 # Non-square pixels
1372 {
1373 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1374 'info_dict': {
1375 'id': '_b-2C3KPAM0',
1376 'ext': 'mp4',
1377 'stretched_ratio': 16 / 9.,
1378 'duration': 85,
1379 'upload_date': '20110310',
1380 'uploader_id': 'AllenMeow',
1381 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1382 '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',
1383 'uploader': '孫ᄋᄅ',
1384 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1385 'playable_in_embed': True,
1386 'channel': '孫ᄋᄅ',
1387 'age_limit': 0,
1388 'tags': 'count:11',
1389 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1390 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1391 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1392 'view_count': int,
1393 'categories': ['People & Blogs'],
1394 'like_count': int,
1395 'live_status': 'not_live',
1396 'availability': 'unlisted',
1397 'channel_follower_count': int
1398 },
1399 },
1400 # url_encoded_fmt_stream_map is empty string
1401 {
1402 'url': 'qEJwOuvDf7I',
1403 'info_dict': {
1404 'id': 'qEJwOuvDf7I',
1405 'ext': 'webm',
1406 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1407 'description': '',
1408 'upload_date': '20150404',
1409 'uploader_id': 'spbelect',
1410 'uploader': 'Наблюдатели Петербурга',
1411 },
1412 'params': {
1413 'skip_download': 'requires avconv',
1414 },
1415 'skip': 'This live event has ended.',
1416 },
1417 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1418 {
1419 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1420 'info_dict': {
1421 'id': 'FIl7x6_3R5Y',
1422 'ext': 'webm',
1423 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1424 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1425 'duration': 220,
1426 'upload_date': '20150625',
1427 'uploader_id': 'dorappi2000',
1428 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1429 'uploader': 'dorappi2000',
1430 'formats': 'mincount:31',
1431 },
1432 'skip': 'not actual anymore',
1433 },
1434 # DASH manifest with segment_list
1435 {
1436 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1437 'md5': '8ce563a1d667b599d21064e982ab9e31',
1438 'info_dict': {
1439 'id': 'CsmdDsKjzN8',
1440 'ext': 'mp4',
1441 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1442 'uploader': 'Airtek',
1443 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1444 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1445 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1446 },
1447 'params': {
1448 'youtube_include_dash_manifest': True,
1449 'format': '135', # bestvideo
1450 },
1451 'skip': 'This live event has ended.',
1452 },
1453 {
1454 # Multifeed videos (multiple cameras), URL is for Main Camera
1455 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1456 'info_dict': {
1457 'id': 'jvGDaLqkpTg',
1458 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1459 'description': 'md5:e03b909557865076822aa169218d6a5d',
1460 },
1461 'playlist': [{
1462 'info_dict': {
1463 'id': 'jvGDaLqkpTg',
1464 'ext': 'mp4',
1465 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1466 'description': 'md5:e03b909557865076822aa169218d6a5d',
1467 'duration': 10643,
1468 'upload_date': '20161111',
1469 'uploader': 'Team PGP',
1470 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1471 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1472 },
1473 }, {
1474 'info_dict': {
1475 'id': '3AKt1R1aDnw',
1476 'ext': 'mp4',
1477 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1478 'description': 'md5:e03b909557865076822aa169218d6a5d',
1479 'duration': 10991,
1480 'upload_date': '20161111',
1481 'uploader': 'Team PGP',
1482 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1483 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1484 },
1485 }, {
1486 'info_dict': {
1487 'id': 'RtAMM00gpVc',
1488 'ext': 'mp4',
1489 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1490 'description': 'md5:e03b909557865076822aa169218d6a5d',
1491 'duration': 10995,
1492 'upload_date': '20161111',
1493 'uploader': 'Team PGP',
1494 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1495 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1496 },
1497 }, {
1498 'info_dict': {
1499 'id': '6N2fdlP3C5U',
1500 'ext': 'mp4',
1501 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1502 'description': 'md5:e03b909557865076822aa169218d6a5d',
1503 'duration': 10990,
1504 'upload_date': '20161111',
1505 'uploader': 'Team PGP',
1506 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1507 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1508 },
1509 }],
1510 'params': {
1511 'skip_download': True,
1512 },
1513 'skip': 'Not multifeed anymore',
1514 },
1515 {
1516 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1517 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1518 'info_dict': {
1519 'id': 'gVfLd0zydlo',
1520 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1521 },
1522 'playlist_count': 2,
1523 'skip': 'Not multifeed anymore',
1524 },
1525 {
1526 'url': 'https://vid.plus/FlRa-iH7PGw',
1527 'only_matching': True,
1528 },
1529 {
1530 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1531 'only_matching': True,
1532 },
1533 {
1534 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1535 # Also tests cut-off URL expansion in video description (see
1536 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1537 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1538 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1539 'info_dict': {
1540 'id': 'lsguqyKfVQg',
1541 'ext': 'mp4',
1542 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1543 'alt_title': 'Dark Walk',
1544 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1545 'duration': 133,
1546 'upload_date': '20151119',
1547 'uploader_id': 'IronSoulElf',
1548 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1549 'uploader': 'IronSoulElf',
1550 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1551 'track': 'Dark Walk',
1552 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1553 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1554 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1555 'categories': ['Film & Animation'],
1556 'view_count': int,
1557 'live_status': 'not_live',
1558 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1559 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1560 'tags': 'count:13',
1561 'availability': 'public',
1562 'channel': 'IronSoulElf',
1563 'playable_in_embed': True,
1564 'like_count': int,
1565 'age_limit': 0,
1566 'channel_follower_count': int
1567 },
1568 'params': {
1569 'skip_download': True,
1570 },
1571 },
1572 {
1573 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1574 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1575 'only_matching': True,
1576 },
1577 {
1578 # Video with yt:stretch=17:0
1579 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1580 'info_dict': {
1581 'id': 'Q39EVAstoRM',
1582 'ext': 'mp4',
1583 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1584 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1585 'upload_date': '20151107',
1586 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1587 'uploader': 'CH GAMER DROID',
1588 },
1589 'params': {
1590 'skip_download': True,
1591 },
1592 'skip': 'This video does not exist.',
1593 },
1594 {
1595 # Video with incomplete 'yt:stretch=16:'
1596 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1597 'only_matching': True,
1598 },
1599 {
1600 # Video licensed under Creative Commons
1601 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1602 'info_dict': {
1603 'id': 'M4gD1WSo5mA',
1604 'ext': 'mp4',
1605 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1606 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1607 'duration': 721,
1608 'upload_date': '20150128',
1609 'uploader_id': 'BerkmanCenter',
1610 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1611 'uploader': 'The Berkman Klein Center for Internet & Society',
1612 'license': 'Creative Commons Attribution license (reuse allowed)',
1613 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1614 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1615 'like_count': int,
1616 'age_limit': 0,
1617 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1618 'channel': 'The Berkman Klein Center for Internet & Society',
1619 'availability': 'public',
1620 'view_count': int,
1621 'categories': ['Education'],
1622 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1623 'live_status': 'not_live',
1624 'playable_in_embed': True,
1625 'channel_follower_count': int
1626 },
1627 'params': {
1628 'skip_download': True,
1629 },
1630 },
1631 {
1632 # Channel-like uploader_url
1633 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1634 'info_dict': {
1635 'id': 'eQcmzGIKrzg',
1636 'ext': 'mp4',
1637 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1638 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1639 'duration': 4060,
1640 'upload_date': '20151120',
1641 'uploader': 'Bernie Sanders',
1642 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1643 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1644 'license': 'Creative Commons Attribution license (reuse allowed)',
1645 'playable_in_embed': True,
1646 'tags': 'count:12',
1647 'like_count': int,
1648 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1649 'age_limit': 0,
1650 'availability': 'public',
1651 'categories': ['News & Politics'],
1652 'channel': 'Bernie Sanders',
1653 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1654 'view_count': int,
1655 'live_status': 'not_live',
1656 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1657 'channel_follower_count': int
1658 },
1659 'params': {
1660 'skip_download': True,
1661 },
1662 },
1663 {
1664 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1665 'only_matching': True,
1666 },
1667 {
1668 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1669 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1670 'only_matching': True,
1671 },
1672 {
1673 # Rental video preview
1674 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1675 'info_dict': {
1676 'id': 'uGpuVWrhIzE',
1677 'ext': 'mp4',
1678 'title': 'Piku - Trailer',
1679 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1680 'upload_date': '20150811',
1681 'uploader': 'FlixMatrix',
1682 'uploader_id': 'FlixMatrixKaravan',
1683 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1684 'license': 'Standard YouTube License',
1685 },
1686 'params': {
1687 'skip_download': True,
1688 },
1689 'skip': 'This video is not available.',
1690 },
1691 {
1692 # YouTube Red video with episode data
1693 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1694 'info_dict': {
1695 'id': 'iqKdEhx-dD4',
1696 'ext': 'mp4',
1697 'title': 'Isolation - Mind Field (Ep 1)',
1698 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1699 'duration': 2085,
1700 'upload_date': '20170118',
1701 'uploader': 'Vsauce',
1702 'uploader_id': 'Vsauce',
1703 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1704 'series': 'Mind Field',
1705 'season_number': 1,
1706 'episode_number': 1,
1707 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1708 'tags': 'count:12',
1709 'view_count': int,
1710 'availability': 'public',
1711 'age_limit': 0,
1712 'channel': 'Vsauce',
1713 'episode': 'Episode 1',
1714 'categories': ['Entertainment'],
1715 'season': 'Season 1',
1716 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1717 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1718 'like_count': int,
1719 'playable_in_embed': True,
1720 'live_status': 'not_live',
1721 'channel_follower_count': int
1722 },
1723 'params': {
1724 'skip_download': True,
1725 },
1726 'expected_warnings': [
1727 'Skipping DASH manifest',
1728 ],
1729 },
1730 {
1731 # The following content has been identified by the YouTube community
1732 # as inappropriate or offensive to some audiences.
1733 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1734 'info_dict': {
1735 'id': '6SJNVb0GnPI',
1736 'ext': 'mp4',
1737 'title': 'Race Differences in Intelligence',
1738 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1739 'duration': 965,
1740 'upload_date': '20140124',
1741 'uploader': 'New Century Foundation',
1742 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1743 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1744 },
1745 'params': {
1746 'skip_download': True,
1747 },
1748 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1749 },
1750 {
1751 # itag 212
1752 'url': '1t24XAntNCY',
1753 'only_matching': True,
1754 },
1755 {
1756 # geo restricted to JP
1757 'url': 'sJL6WA-aGkQ',
1758 'only_matching': True,
1759 },
1760 {
1761 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1762 'only_matching': True,
1763 },
1764 {
1765 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1766 'only_matching': True,
1767 },
1768 {
1769 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1770 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1771 'only_matching': True,
1772 },
1773 {
1774 # DRM protected
1775 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1776 'only_matching': True,
1777 },
1778 {
1779 # Video with unsupported adaptive stream type formats
1780 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1781 'info_dict': {
1782 'id': 'Z4Vy8R84T1U',
1783 'ext': 'mp4',
1784 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1785 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1786 'duration': 433,
1787 'upload_date': '20130923',
1788 'uploader': 'Amelia Putri Harwita',
1789 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1790 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1791 'formats': 'maxcount:10',
1792 },
1793 'params': {
1794 'skip_download': True,
1795 'youtube_include_dash_manifest': False,
1796 },
1797 'skip': 'not actual anymore',
1798 },
1799 {
1800 # Youtube Music Auto-generated description
1801 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1802 'info_dict': {
1803 'id': 'MgNrAu2pzNs',
1804 'ext': 'mp4',
1805 'title': 'Voyeur Girl',
1806 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1807 'upload_date': '20190312',
1808 'uploader': 'Stephen - Topic',
1809 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1810 'artist': 'Stephen',
1811 'track': 'Voyeur Girl',
1812 'album': 'it\'s too much love to know my dear',
1813 'release_date': '20190313',
1814 'release_year': 2019,
1815 'alt_title': 'Voyeur Girl',
1816 'view_count': int,
1817 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1818 'playable_in_embed': True,
1819 'like_count': int,
1820 'categories': ['Music'],
1821 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1822 'channel': 'Stephen',
1823 'availability': 'public',
1824 'creator': 'Stephen',
1825 'duration': 169,
1826 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1827 'age_limit': 0,
1828 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1829 'tags': 'count:11',
1830 'live_status': 'not_live',
1831 'channel_follower_count': int
1832 },
1833 'params': {
1834 'skip_download': True,
1835 },
1836 },
1837 {
1838 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1839 'only_matching': True,
1840 },
1841 {
1842 # invalid -> valid video id redirection
1843 'url': 'DJztXj2GPfl',
1844 'info_dict': {
1845 'id': 'DJztXj2GPfk',
1846 'ext': 'mp4',
1847 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1848 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1849 'upload_date': '20090125',
1850 'uploader': 'Prochorowka',
1851 'uploader_id': 'Prochorowka',
1852 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1853 'artist': 'Panjabi MC',
1854 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1855 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1856 },
1857 'params': {
1858 'skip_download': True,
1859 },
1860 'skip': 'Video unavailable',
1861 },
1862 {
1863 # empty description results in an empty string
1864 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1865 'info_dict': {
1866 'id': 'x41yOUIvK2k',
1867 'ext': 'mp4',
1868 'title': 'IMG 3456',
1869 'description': '',
1870 'upload_date': '20170613',
1871 'uploader_id': 'ElevageOrVert',
1872 'uploader': 'ElevageOrVert',
1873 'view_count': int,
1874 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1875 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1876 'like_count': int,
1877 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1878 'tags': [],
1879 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1880 'availability': 'public',
1881 'age_limit': 0,
1882 'categories': ['Pets & Animals'],
1883 'duration': 7,
1884 'playable_in_embed': True,
1885 'live_status': 'not_live',
1886 'channel': 'ElevageOrVert',
1887 'channel_follower_count': int
1888 },
1889 'params': {
1890 'skip_download': True,
1891 },
1892 },
1893 {
1894 # with '};' inside yt initial data (see [1])
1895 # see [2] for an example with '};' inside ytInitialPlayerResponse
1896 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1897 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1898 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1899 'info_dict': {
1900 'id': 'CHqg6qOn4no',
1901 'ext': 'mp4',
1902 'title': 'Part 77 Sort a list of simple types in c#',
1903 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1904 'upload_date': '20130831',
1905 'uploader_id': 'kudvenkat',
1906 'uploader': 'kudvenkat',
1907 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1908 'like_count': int,
1909 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1910 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1911 'live_status': 'not_live',
1912 'categories': ['Education'],
1913 'availability': 'public',
1914 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1915 'tags': 'count:12',
1916 'playable_in_embed': True,
1917 'age_limit': 0,
1918 'view_count': int,
1919 'duration': 522,
1920 'channel': 'kudvenkat',
1921 'channel_follower_count': int
1922 },
1923 'params': {
1924 'skip_download': True,
1925 },
1926 },
1927 {
1928 # another example of '};' in ytInitialData
1929 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1930 'only_matching': True,
1931 },
1932 {
1933 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1934 'only_matching': True,
1935 },
1936 {
1937 # https://github.com/ytdl-org/youtube-dl/pull/28094
1938 'url': 'OtqTfy26tG0',
1939 'info_dict': {
1940 'id': 'OtqTfy26tG0',
1941 'ext': 'mp4',
1942 'title': 'Burn Out',
1943 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1944 'upload_date': '20141120',
1945 'uploader': 'The Cinematic Orchestra - Topic',
1946 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1947 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1948 'artist': 'The Cinematic Orchestra',
1949 'track': 'Burn Out',
1950 'album': 'Every Day',
1951 'like_count': int,
1952 'live_status': 'not_live',
1953 'alt_title': 'Burn Out',
1954 'duration': 614,
1955 'age_limit': 0,
1956 'view_count': int,
1957 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1958 'creator': 'The Cinematic Orchestra',
1959 'channel': 'The Cinematic Orchestra',
1960 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1961 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1962 'availability': 'public',
1963 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1964 'categories': ['Music'],
1965 'playable_in_embed': True,
1966 'channel_follower_count': int
1967 },
1968 'params': {
1969 'skip_download': True,
1970 },
1971 },
1972 {
1973 # controversial video, only works with bpctr when authenticated with cookies
1974 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1975 'only_matching': True,
1976 },
1977 {
1978 # controversial video, requires bpctr/contentCheckOk
1979 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1980 'info_dict': {
1981 'id': 'SZJvDhaSDnc',
1982 'ext': 'mp4',
1983 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1984 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1985 'uploader': 'CBS Mornings',
1986 'uploader_id': 'CBSThisMorning',
1987 'upload_date': '20140716',
1988 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
1989 'duration': 170,
1990 'categories': ['News & Politics'],
1991 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
1992 'view_count': int,
1993 'channel': 'CBS Mornings',
1994 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
1995 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
1996 'age_limit': 18,
1997 'availability': 'needs_auth',
1998 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
1999 'like_count': int,
2000 'live_status': 'not_live',
2001 'playable_in_embed': True,
2002 'channel_follower_count': int
2003 }
2004 },
2005 {
2006 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2007 'url': 'cBvYw8_A0vQ',
2008 'info_dict': {
2009 'id': 'cBvYw8_A0vQ',
2010 'ext': 'mp4',
2011 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2012 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2013 'upload_date': '20201120',
2014 'uploader': 'Walk around Japan',
2015 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2016 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2017 'duration': 1456,
2018 'categories': ['Travel & Events'],
2019 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2020 'view_count': int,
2021 'channel': 'Walk around Japan',
2022 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2023 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2024 'age_limit': 0,
2025 'availability': 'public',
2026 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2027 'live_status': 'not_live',
2028 'playable_in_embed': True,
2029 'channel_follower_count': int
2030 },
2031 'params': {
2032 'skip_download': True,
2033 },
2034 }, {
2035 # Has multiple audio streams
2036 'url': 'WaOKSUlf4TM',
2037 'only_matching': True
2038 }, {
2039 # Requires Premium: has format 141 when requested using YTM url
2040 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2041 'only_matching': True
2042 }, {
2043 # multiple subtitles with same lang_code
2044 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2045 'only_matching': True,
2046 }, {
2047 # Force use android client fallback
2048 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2049 'info_dict': {
2050 'id': 'YOelRv7fMxY',
2051 'title': 'DIGGING A SECRET TUNNEL Part 1',
2052 'ext': '3gp',
2053 'upload_date': '20210624',
2054 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2055 'uploader': 'colinfurze',
2056 'uploader_id': 'colinfurze',
2057 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2058 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2059 'duration': 596,
2060 'categories': ['Entertainment'],
2061 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2062 'view_count': int,
2063 'channel': 'colinfurze',
2064 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2065 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2066 'age_limit': 0,
2067 'availability': 'public',
2068 'like_count': int,
2069 'live_status': 'not_live',
2070 'playable_in_embed': True,
2071 'channel_follower_count': int
2072 },
2073 'params': {
2074 'format': '17', # 3gp format available on android
2075 'extractor_args': {'youtube': {'player_client': ['android']}},
2076 },
2077 },
2078 {
2079 # Skip download of additional client configs (remix client config in this case)
2080 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2081 'only_matching': True,
2082 'params': {
2083 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2084 },
2085 }, {
2086 # shorts
2087 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2088 'only_matching': True,
2089 }, {
2090 'note': 'Storyboards',
2091 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2092 'info_dict': {
2093 'id': '5KLPxDtMqe8',
2094 'ext': 'mhtml',
2095 'format_id': 'sb0',
2096 'title': 'Your Brain is Plastic',
2097 'uploader_id': 'scishow',
2098 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2099 'upload_date': '20140324',
2100 'uploader': 'SciShow',
2101 'like_count': int,
2102 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2103 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2104 'view_count': int,
2105 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2106 'playable_in_embed': True,
2107 'tags': 'count:12',
2108 'uploader_url': 'http://www.youtube.com/user/scishow',
2109 'availability': 'public',
2110 'channel': 'SciShow',
2111 'live_status': 'not_live',
2112 'duration': 248,
2113 'categories': ['Education'],
2114 'age_limit': 0,
2115 'channel_follower_count': int
2116 }, 'params': {'format': 'mhtml', 'skip_download': True}
2117 }, {
2118 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2119 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2120 'info_dict': {
2121 'id': '2NUZ8W2llS4',
2122 'ext': 'mp4',
2123 'title': 'The NP that test your phone performance 🙂',
2124 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2125 'uploader': 'Leon Nguyen',
2126 'uploader_id': 'VNSXIII',
2127 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2128 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2129 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2130 'duration': 21,
2131 'view_count': int,
2132 'age_limit': 0,
2133 'categories': ['Gaming'],
2134 'tags': 'count:23',
2135 'playable_in_embed': True,
2136 'live_status': 'not_live',
2137 'upload_date': '20220103',
2138 'like_count': int,
2139 'availability': 'public',
2140 'channel': 'Leon Nguyen',
2141 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2142 'channel_follower_count': int
2143 }
2144 }, {
2145 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2146 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2147 'info_dict': {
2148 'id': 'mzZzzBU6lrM',
2149 'ext': 'mp4',
2150 'title': 'I Met GeorgeNotFound In Real Life...',
2151 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2152 'uploader': 'Quackity',
2153 'uploader_id': 'QuackityHQ',
2154 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2155 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2156 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2157 'duration': 955,
2158 'view_count': int,
2159 'age_limit': 0,
2160 'categories': ['Entertainment'],
2161 'tags': 'count:26',
2162 'playable_in_embed': True,
2163 'live_status': 'not_live',
2164 'release_timestamp': 1641172509,
2165 'release_date': '20220103',
2166 'upload_date': '20220103',
2167 'like_count': int,
2168 'availability': 'public',
2169 'channel': 'Quackity',
2170 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2171 'channel_follower_count': int
2172 }
2173 },
2174 { # continuous livestream. Microformat upload date should be preferred.
2175 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2176 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2177 'info_dict': {
2178 'id': 'kgx4WGK0oNU',
2179 '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}',
2180 'ext': 'mp4',
2181 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2182 'availability': 'public',
2183 'age_limit': 0,
2184 'release_timestamp': 1637975704,
2185 'upload_date': '20210619',
2186 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2187 'live_status': 'is_live',
2188 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2189 'uploader': '阿鲍Abao',
2190 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2191 'channel': 'Abao in Tokyo',
2192 'channel_follower_count': int,
2193 'release_date': '20211127',
2194 'tags': 'count:39',
2195 'categories': ['People & Blogs'],
2196 'like_count': int,
2197 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2198 'view_count': int,
2199 'playable_in_embed': True,
2200 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2201 },
2202 'params': {'skip_download': True}
2203 }, {
2204 # Story. Requires specific player params to work.
2205 # Note: stories get removed after some period of time
2206 'url': 'https://www.youtube.com/watch?v=yN3x1t3sieA',
2207 'info_dict': {
2208 'id': 'yN3x1t3sieA',
2209 'ext': 'mp4',
2210 'uploader': 'Linus Tech Tips',
2211 'duration': 13,
2212 'channel': 'Linus Tech Tips',
2213 'playable_in_embed': True,
2214 'tags': [],
2215 'age_limit': 0,
2216 'uploader_url': 'http://www.youtube.com/user/LinusTechTips',
2217 'upload_date': '20220402',
2218 'thumbnail': 'https://i.ytimg.com/vi_webp/yN3x1t3sieA/maxresdefault.webp',
2219 'title': 'Story',
2220 'live_status': 'not_live',
2221 'uploader_id': 'LinusTechTips',
2222 'view_count': int,
2223 'description': '',
2224 'channel_id': 'UCXuqSBlHAE6Xw-yeJA0Tunw',
2225 'categories': ['Science & Technology'],
2226 'channel_url': 'https://www.youtube.com/channel/UCXuqSBlHAE6Xw-yeJA0Tunw',
2227 'availability': 'unlisted',
2228 }
2229 }
2230 ]
2231
2232 @classmethod
2233 def suitable(cls, url):
2234 from ..utils import parse_qs
2235
2236 qs = parse_qs(url)
2237 if qs.get('list', [None])[0]:
2238 return False
2239 return super().suitable(url)
2240
2241 def __init__(self, *args, **kwargs):
2242 super().__init__(*args, **kwargs)
2243 self._code_cache = {}
2244 self._player_cache = {}
2245
2246 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2247 lock = threading.Lock()
2248
2249 is_live = True
2250 start_time = time.time()
2251 formats = [f for f in formats if f.get('is_from_start')]
2252
2253 def refetch_manifest(format_id, delay):
2254 nonlocal formats, start_time, is_live
2255 if time.time() <= start_time + delay:
2256 return
2257
2258 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2259 video_details = traverse_obj(
2260 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2261 microformats = traverse_obj(
2262 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2263 expected_type=dict, default=[])
2264 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2265 start_time = time.time()
2266
2267 def mpd_feed(format_id, delay):
2268 """
2269 @returns (manifest_url, manifest_stream_number, is_live) or None
2270 """
2271 with lock:
2272 refetch_manifest(format_id, delay)
2273
2274 f = next((f for f in formats if f['format_id'] == format_id), None)
2275 if not f:
2276 if not is_live:
2277 self.to_screen(f'{video_id}: Video is no longer live')
2278 else:
2279 self.report_warning(
2280 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2281 return None
2282 return f['manifest_url'], f['manifest_stream_number'], is_live
2283
2284 for f in formats:
2285 f['is_live'] = True
2286 f['protocol'] = 'http_dash_segments_generator'
2287 f['fragments'] = functools.partial(
2288 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2289
2290 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2291 FETCH_SPAN, MAX_DURATION = 5, 432000
2292
2293 mpd_url, stream_number, is_live = None, None, True
2294
2295 begin_index = 0
2296 download_start_time = ctx.get('start') or time.time()
2297
2298 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2299 if lack_early_segments:
2300 self.report_warning(bug_reports_message(
2301 'Starting download from the last 120 hours of the live stream since '
2302 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2303 lack_early_segments = True
2304
2305 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2306 fragments, fragment_base_url = None, None
2307
2308 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2309 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2310 # Obtain from MPD's maximum seq value
2311 old_mpd_url = mpd_url
2312 last_error = ctx.pop('last_error', None)
2313 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2314 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2315 or (mpd_url, stream_number, False))
2316 if not refresh_sequence:
2317 if expire_fast and not is_live:
2318 return False, last_seq
2319 elif old_mpd_url == mpd_url:
2320 return True, last_seq
2321 try:
2322 fmts, _ = self._extract_mpd_formats_and_subtitles(
2323 mpd_url, None, note=False, errnote=False, fatal=False)
2324 except ExtractorError:
2325 fmts = None
2326 if not fmts:
2327 no_fragment_score += 2
2328 return False, last_seq
2329 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2330 fragments = fmt_info['fragments']
2331 fragment_base_url = fmt_info['fragment_base_url']
2332 assert fragment_base_url
2333
2334 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2335 return True, _last_seq
2336
2337 while is_live:
2338 fetch_time = time.time()
2339 if no_fragment_score > 30:
2340 return
2341 if last_segment_url:
2342 # Obtain from "X-Head-Seqnum" header value from each segment
2343 try:
2344 urlh = self._request_webpage(
2345 last_segment_url, None, note=False, errnote=False, fatal=False)
2346 except ExtractorError:
2347 urlh = None
2348 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2349 if last_seq is None:
2350 no_fragment_score += 2
2351 last_segment_url = None
2352 continue
2353 else:
2354 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2355 no_fragment_score += 2
2356 if not should_continue:
2357 continue
2358
2359 if known_idx > last_seq:
2360 last_segment_url = None
2361 continue
2362
2363 last_seq += 1
2364
2365 if begin_index < 0 and known_idx < 0:
2366 # skip from the start when it's negative value
2367 known_idx = last_seq + begin_index
2368 if lack_early_segments:
2369 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2370 try:
2371 for idx in range(known_idx, last_seq):
2372 # do not update sequence here or you'll get skipped some part of it
2373 should_continue, _ = _extract_sequence_from_mpd(False, False)
2374 if not should_continue:
2375 known_idx = idx - 1
2376 raise ExtractorError('breaking out of outer loop')
2377 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2378 yield {
2379 'url': last_segment_url,
2380 }
2381 if known_idx == last_seq:
2382 no_fragment_score += 5
2383 else:
2384 no_fragment_score = 0
2385 known_idx = last_seq
2386 except ExtractorError:
2387 continue
2388
2389 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2390
2391 def _extract_player_url(self, *ytcfgs, webpage=None):
2392 player_url = traverse_obj(
2393 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2394 get_all=False, expected_type=compat_str)
2395 if not player_url:
2396 return
2397 return urljoin('https://www.youtube.com', player_url)
2398
2399 def _download_player_url(self, video_id, fatal=False):
2400 res = self._download_webpage(
2401 'https://www.youtube.com/iframe_api',
2402 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2403 if res:
2404 player_version = self._search_regex(
2405 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2406 if player_version:
2407 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2408
2409 def _signature_cache_id(self, example_sig):
2410 """ Return a string representation of a signature """
2411 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2412
2413 @classmethod
2414 def _extract_player_info(cls, player_url):
2415 for player_re in cls._PLAYER_INFO_RE:
2416 id_m = re.search(player_re, player_url)
2417 if id_m:
2418 break
2419 else:
2420 raise ExtractorError('Cannot identify player %r' % player_url)
2421 return id_m.group('id')
2422
2423 def _load_player(self, video_id, player_url, fatal=True):
2424 player_id = self._extract_player_info(player_url)
2425 if player_id not in self._code_cache:
2426 code = self._download_webpage(
2427 player_url, video_id, fatal=fatal,
2428 note='Downloading player ' + player_id,
2429 errnote='Download of %s failed' % player_url)
2430 if code:
2431 self._code_cache[player_id] = code
2432 return self._code_cache.get(player_id)
2433
2434 def _extract_signature_function(self, video_id, player_url, example_sig):
2435 player_id = self._extract_player_info(player_url)
2436
2437 # Read from filesystem cache
2438 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2439 assert os.path.basename(func_id) == func_id
2440
2441 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2442 if cache_spec is not None:
2443 return lambda s: ''.join(s[i] for i in cache_spec)
2444
2445 code = self._load_player(video_id, player_url)
2446 if code:
2447 res = self._parse_sig_js(code)
2448
2449 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2450 cache_res = res(test_string)
2451 cache_spec = [ord(c) for c in cache_res]
2452
2453 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2454 return res
2455
2456 def _print_sig_code(self, func, example_sig):
2457 if not self.get_param('youtube_print_sig_code'):
2458 return
2459
2460 def gen_sig_code(idxs):
2461 def _genslice(start, end, step):
2462 starts = '' if start == 0 else str(start)
2463 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2464 steps = '' if step == 1 else (':%d' % step)
2465 return f's[{starts}{ends}{steps}]'
2466
2467 step = None
2468 # Quelch pyflakes warnings - start will be set when step is set
2469 start = '(Never used)'
2470 for i, prev in zip(idxs[1:], idxs[:-1]):
2471 if step is not None:
2472 if i - prev == step:
2473 continue
2474 yield _genslice(start, prev, step)
2475 step = None
2476 continue
2477 if i - prev in [-1, 1]:
2478 step = i - prev
2479 start = prev
2480 continue
2481 else:
2482 yield 's[%d]' % prev
2483 if step is None:
2484 yield 's[%d]' % i
2485 else:
2486 yield _genslice(start, i, step)
2487
2488 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2489 cache_res = func(test_string)
2490 cache_spec = [ord(c) for c in cache_res]
2491 expr_code = ' + '.join(gen_sig_code(cache_spec))
2492 signature_id_tuple = '(%s)' % (
2493 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2494 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2495 ' return %s\n') % (signature_id_tuple, expr_code)
2496 self.to_screen('Extracted signature function:\n' + code)
2497
2498 def _parse_sig_js(self, jscode):
2499 funcname = self._search_regex(
2500 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2501 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2502 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2503 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2504 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}\(a,\d+\)',
2505 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\(\s*""\s*\)',
2506 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2507 # Obsolete patterns
2508 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2509 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2510 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2511 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2512 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2513 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2514 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2515 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2516 jscode, 'Initial JS player signature function name', group='sig')
2517
2518 jsi = JSInterpreter(jscode)
2519 initial_function = jsi.extract_function(funcname)
2520 return lambda s: initial_function([s])
2521
2522 def _decrypt_signature(self, s, video_id, player_url):
2523 """Turn the encrypted s field into a working signature"""
2524
2525 if player_url is None:
2526 raise ExtractorError('Cannot decrypt signature without player_url')
2527
2528 try:
2529 player_id = (player_url, self._signature_cache_id(s))
2530 if player_id not in self._player_cache:
2531 func = self._extract_signature_function(
2532 video_id, player_url, s
2533 )
2534 self._player_cache[player_id] = func
2535 func = self._player_cache[player_id]
2536 self._print_sig_code(func, s)
2537 return func(s)
2538 except Exception as e:
2539 raise ExtractorError('Signature extraction failed: ' + traceback.format_exc(), cause=e)
2540
2541 def _decrypt_nsig(self, s, video_id, player_url):
2542 """Turn the encrypted n field into a working signature"""
2543 if player_url is None:
2544 raise ExtractorError('Cannot decrypt nsig without player_url')
2545 player_url = urljoin('https://www.youtube.com', player_url)
2546
2547 sig_id = ('nsig_value', s)
2548 if sig_id in self._player_cache:
2549 return self._player_cache[sig_id]
2550
2551 try:
2552 player_id = ('nsig', player_url)
2553 if player_id not in self._player_cache:
2554 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2555 func = self._player_cache[player_id]
2556 self._player_cache[sig_id] = func(s)
2557 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2558 return self._player_cache[sig_id]
2559 except Exception as e:
2560 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2561
2562 def _extract_n_function_name(self, jscode):
2563 nfunc, idx = self._search_regex(
2564 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2565 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2566 if not idx:
2567 return nfunc
2568 return json.loads(js_to_json(self._search_regex(
2569 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2570 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2571
2572 def _extract_n_function(self, video_id, player_url):
2573 player_id = self._extract_player_info(player_url)
2574 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2575
2576 if func_code:
2577 jsi = JSInterpreter(func_code)
2578 else:
2579 jscode = self._load_player(video_id, player_url)
2580 funcname = self._extract_n_function_name(jscode)
2581 jsi = JSInterpreter(jscode)
2582 func_code = jsi.extract_function_code(funcname)
2583 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2584
2585 if self.get_param('youtube_print_sig_code'):
2586 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2587
2588 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2589
2590 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2591 """
2592 Extract signatureTimestamp (sts)
2593 Required to tell API what sig/player version is in use.
2594 """
2595 sts = None
2596 if isinstance(ytcfg, dict):
2597 sts = int_or_none(ytcfg.get('STS'))
2598
2599 if not sts:
2600 # Attempt to extract from player
2601 if player_url is None:
2602 error_msg = 'Cannot extract signature timestamp without player_url.'
2603 if fatal:
2604 raise ExtractorError(error_msg)
2605 self.report_warning(error_msg)
2606 return
2607 code = self._load_player(video_id, player_url, fatal=fatal)
2608 if code:
2609 sts = int_or_none(self._search_regex(
2610 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2611 'JS player signature timestamp', group='sts', fatal=fatal))
2612 return sts
2613
2614 def _mark_watched(self, video_id, player_responses):
2615 playback_url = get_first(
2616 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2617 expected_type=url_or_none)
2618 if not playback_url:
2619 self.report_warning('Unable to mark watched')
2620 return
2621 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2622 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2623
2624 # cpn generation algorithm is reverse engineered from base.js.
2625 # In fact it works even with dummy cpn.
2626 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2627 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2628
2629 qs.update({
2630 'ver': ['2'],
2631 'cpn': [cpn],
2632 })
2633 playback_url = compat_urlparse.urlunparse(
2634 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2635
2636 self._download_webpage(
2637 playback_url, video_id, 'Marking watched',
2638 'Unable to mark watched', fatal=False)
2639
2640 @staticmethod
2641 def _extract_urls(webpage):
2642 # Embedded YouTube player
2643 entries = [
2644 unescapeHTML(mobj.group('url'))
2645 for mobj in re.finditer(r'''(?x)
2646 (?:
2647 <iframe[^>]+?src=|
2648 data-video-url=|
2649 <embed[^>]+?src=|
2650 embedSWF\(?:\s*|
2651 <object[^>]+data=|
2652 new\s+SWFObject\(
2653 )
2654 (["\'])
2655 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2656 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2657 \1''', webpage)]
2658
2659 # lazyYT YouTube embed
2660 entries.extend(list(map(
2661 unescapeHTML,
2662 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2663
2664 # Wordpress "YouTube Video Importer" plugin
2665 matches = re.findall(r'''(?x)<div[^>]+
2666 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2667 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2668 entries.extend(m[-1] for m in matches)
2669
2670 return entries
2671
2672 @staticmethod
2673 def _extract_url(webpage):
2674 urls = YoutubeIE._extract_urls(webpage)
2675 return urls[0] if urls else None
2676
2677 @classmethod
2678 def extract_id(cls, url):
2679 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2680 if mobj is None:
2681 raise ExtractorError('Invalid URL: %s' % url)
2682 return mobj.group('id')
2683
2684 def _extract_chapters_from_json(self, data, duration):
2685 chapter_list = traverse_obj(
2686 data, (
2687 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2688 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2689 ), expected_type=list)
2690
2691 return self._extract_chapters(
2692 chapter_list,
2693 chapter_time=lambda chapter: float_or_none(
2694 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2695 chapter_title=lambda chapter: traverse_obj(
2696 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2697 duration=duration)
2698
2699 def _extract_chapters_from_engagement_panel(self, data, duration):
2700 content_list = traverse_obj(
2701 data,
2702 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2703 expected_type=list, default=[])
2704 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2705 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2706
2707 return next((
2708 filter(None, (
2709 self._extract_chapters(
2710 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2711 chapter_time, chapter_title, duration)
2712 for contents in content_list
2713 ))), [])
2714
2715 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2716 chapters = []
2717 last_chapter = {'start_time': 0}
2718 for idx, chapter in enumerate(chapter_list or []):
2719 title = chapter_title(chapter)
2720 start_time = chapter_time(chapter)
2721 if start_time is None:
2722 continue
2723 last_chapter['end_time'] = start_time
2724 if start_time < last_chapter['start_time']:
2725 if idx == 1:
2726 chapters.pop()
2727 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2728 else:
2729 self.report_warning(f'Invalid start time for chapter "{title}"')
2730 continue
2731 last_chapter = {'start_time': start_time, 'title': title}
2732 chapters.append(last_chapter)
2733 last_chapter['end_time'] = duration
2734 return chapters
2735
2736 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2737 return self._parse_json(self._search_regex(
2738 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
2739 regex), webpage, name, default='{}'), video_id, fatal=False)
2740
2741 def _extract_comment(self, comment_renderer, parent=None):
2742 comment_id = comment_renderer.get('commentId')
2743 if not comment_id:
2744 return
2745
2746 text = self._get_text(comment_renderer, 'contentText')
2747
2748 # note: timestamp is an estimate calculated from the current time and time_text
2749 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2750 author = self._get_text(comment_renderer, 'authorText')
2751 author_id = try_get(comment_renderer,
2752 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2753
2754 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2755 lambda x: x['likeCount']), compat_str)) or 0
2756 author_thumbnail = try_get(comment_renderer,
2757 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2758
2759 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2760 is_favorited = 'creatorHeart' in (try_get(
2761 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2762 return {
2763 'id': comment_id,
2764 'text': text,
2765 'timestamp': timestamp,
2766 'time_text': time_text,
2767 'like_count': votes,
2768 'is_favorited': is_favorited,
2769 'author': author,
2770 'author_id': author_id,
2771 'author_thumbnail': author_thumbnail,
2772 'author_is_uploader': author_is_uploader,
2773 'parent': parent or 'root'
2774 }
2775
2776 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2777
2778 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2779
2780 def extract_header(contents):
2781 _continuation = None
2782 for content in contents:
2783 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2784 expected_comment_count = self._get_count(
2785 comments_header_renderer, 'countText', 'commentsCount')
2786
2787 if expected_comment_count:
2788 tracker['est_total'] = expected_comment_count
2789 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2790 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2791
2792 sort_menu_item = try_get(
2793 comments_header_renderer,
2794 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2795 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2796
2797 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2798 if not _continuation:
2799 continue
2800
2801 sort_text = str_or_none(sort_menu_item.get('title'))
2802 if not sort_text:
2803 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2804 self.to_screen('Sorting comments by %s' % sort_text.lower())
2805 break
2806 return _continuation
2807
2808 def extract_thread(contents):
2809 if not parent:
2810 tracker['current_page_thread'] = 0
2811 for content in contents:
2812 if not parent and tracker['total_parent_comments'] >= max_parents:
2813 yield
2814 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2815 comment_renderer = get_first(
2816 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2817 expected_type=dict, default={})
2818
2819 comment = self._extract_comment(comment_renderer, parent)
2820 if not comment:
2821 continue
2822
2823 tracker['running_total'] += 1
2824 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2825 yield comment
2826
2827 # Attempt to get the replies
2828 comment_replies_renderer = try_get(
2829 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2830
2831 if comment_replies_renderer:
2832 tracker['current_page_thread'] += 1
2833 comment_entries_iter = self._comment_entries(
2834 comment_replies_renderer, ytcfg, video_id,
2835 parent=comment.get('id'), tracker=tracker)
2836 yield from itertools.islice(comment_entries_iter, min(
2837 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2838
2839 # Keeps track of counts across recursive calls
2840 if not tracker:
2841 tracker = dict(
2842 running_total=0,
2843 est_total=0,
2844 current_page_thread=0,
2845 total_parent_comments=0,
2846 total_reply_comments=0)
2847
2848 # TODO: Deprecated
2849 # YouTube comments have a max depth of 2
2850 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2851 if max_depth:
2852 self._downloader.deprecation_warning(
2853 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2854 if max_depth == 1 and parent:
2855 return
2856
2857 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2858 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2859
2860 continuation = self._extract_continuation(root_continuation_data)
2861
2862 response = None
2863 is_forced_continuation = False
2864 is_first_continuation = parent is None
2865 if is_first_continuation and not continuation:
2866 # Sometimes you can get comments by generating the continuation yourself,
2867 # even if YouTube initially reports them being disabled - e.g. stories comments.
2868 # Note: if the comment section is actually disabled, YouTube may return a response with
2869 # required check_get_keys missing. So we will disable that check initially in this case.
2870 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2871 is_forced_continuation = True
2872
2873 for page_num in itertools.count(0):
2874 if not continuation:
2875 break
2876 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2877 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2878 if page_num == 0:
2879 if is_first_continuation:
2880 note_prefix = 'Downloading comment section API JSON'
2881 else:
2882 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2883 tracker['current_page_thread'], comment_prog_str)
2884 else:
2885 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2886 ' ' if parent else '', ' replies' if parent else '',
2887 page_num, comment_prog_str)
2888
2889 response = self._extract_response(
2890 item_id=None, query=continuation,
2891 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2892 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2893 is_forced_continuation = False
2894 continuation_contents = traverse_obj(
2895 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2896
2897 continuation = None
2898 for continuation_section in continuation_contents:
2899 continuation_items = traverse_obj(
2900 continuation_section,
2901 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2902 get_all=False, expected_type=list) or []
2903 if is_first_continuation:
2904 continuation = extract_header(continuation_items)
2905 is_first_continuation = False
2906 if continuation:
2907 break
2908 continue
2909
2910 for entry in extract_thread(continuation_items):
2911 if not entry:
2912 return
2913 yield entry
2914 continuation = self._extract_continuation({'contents': continuation_items})
2915 if continuation:
2916 break
2917
2918 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2919 if message and not parent and tracker['running_total'] == 0:
2920 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2921
2922 @staticmethod
2923 def _generate_comment_continuation(video_id):
2924 """
2925 Generates initial comment section continuation token from given video id
2926 """
2927 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2928 return base64.b64encode(token.encode()).decode()
2929
2930 def _get_comments(self, ytcfg, video_id, contents, webpage):
2931 """Entry for comment extraction"""
2932 def _real_comment_extract(contents):
2933 renderer = next((
2934 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2935 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2936 yield from self._comment_entries(renderer, ytcfg, video_id)
2937
2938 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2939 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2940
2941 @staticmethod
2942 def _get_checkok_params():
2943 return {'contentCheckOk': True, 'racyCheckOk': True}
2944
2945 @classmethod
2946 def _generate_player_context(cls, sts=None):
2947 context = {
2948 'html5Preference': 'HTML5_PREF_WANTS',
2949 }
2950 if sts is not None:
2951 context['signatureTimestamp'] = sts
2952 return {
2953 'playbackContext': {
2954 'contentPlaybackContext': context
2955 },
2956 **cls._get_checkok_params()
2957 }
2958
2959 @staticmethod
2960 def _is_agegated(player_response):
2961 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2962 return True
2963
2964 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2965 AGE_GATE_REASONS = (
2966 'confirm your age', 'age-restricted', 'inappropriate', # reason
2967 'age_verification_required', 'age_check_required', # status
2968 )
2969 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2970
2971 @staticmethod
2972 def _is_unplayable(player_response):
2973 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2974
2975 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2976
2977 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2978 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2979 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2980 headers = self.generate_api_headers(
2981 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
2982
2983 yt_query = {
2984 'videoId': video_id,
2985 'params': '8AEB' # enable stories
2986 }
2987 yt_query.update(self._generate_player_context(sts))
2988 return self._extract_response(
2989 item_id=video_id, ep='player', query=yt_query,
2990 ytcfg=player_ytcfg, headers=headers, fatal=True,
2991 default_client=client,
2992 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
2993 ) or None
2994
2995 def _get_requested_clients(self, url, smuggled_data):
2996 requested_clients = []
2997 default = ['android', 'web']
2998 allowed_clients = sorted(
2999 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3000 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3001 for client in self._configuration_arg('player_client'):
3002 if client in allowed_clients:
3003 requested_clients.append(client)
3004 elif client == 'default':
3005 requested_clients.extend(default)
3006 elif client == 'all':
3007 requested_clients.extend(allowed_clients)
3008 else:
3009 self.report_warning(f'Skipping unsupported client {client}')
3010 if not requested_clients:
3011 requested_clients = default
3012
3013 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3014 requested_clients.extend(
3015 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3016
3017 return orderedSet(requested_clients)
3018
3019 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3020 initial_pr = None
3021 if webpage:
3022 initial_pr = self._extract_yt_initial_variable(
3023 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
3024 video_id, 'initial player response')
3025
3026 all_clients = set(clients)
3027 clients = clients[::-1]
3028 prs = []
3029
3030 def append_client(*client_names):
3031 """ Append the first client name that exists but not already used """
3032 for client_name in client_names:
3033 actual_client = _split_innertube_client(client_name)[0]
3034 if actual_client in INNERTUBE_CLIENTS:
3035 if actual_client not in all_clients:
3036 clients.append(client_name)
3037 all_clients.add(actual_client)
3038 return
3039
3040 # Android player_response does not have microFormats which are needed for
3041 # extraction of some data. So we return the initial_pr with formats
3042 # stripped out even if not requested by the user
3043 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3044 if initial_pr:
3045 pr = dict(initial_pr)
3046 pr['streamingData'] = None
3047 prs.append(pr)
3048
3049 last_error = None
3050 tried_iframe_fallback = False
3051 player_url = None
3052 while clients:
3053 client, base_client, variant = _split_innertube_client(clients.pop())
3054 player_ytcfg = master_ytcfg if client == 'web' else {}
3055 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3056 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3057
3058 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3059 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3060 if 'js' in self._configuration_arg('player_skip'):
3061 require_js_player = False
3062 player_url = None
3063
3064 if not player_url and not tried_iframe_fallback and require_js_player:
3065 player_url = self._download_player_url(video_id)
3066 tried_iframe_fallback = True
3067
3068 try:
3069 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3070 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3071 except ExtractorError as e:
3072 if last_error:
3073 self.report_warning(last_error)
3074 last_error = e
3075 continue
3076
3077 if pr:
3078 prs.append(pr)
3079
3080 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3081 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3082 append_client(f'{base_client}_creator')
3083 elif self._is_agegated(pr):
3084 if variant == 'tv_embedded':
3085 append_client(f'{base_client}_embedded')
3086 elif not variant:
3087 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3088
3089 if last_error:
3090 if not len(prs):
3091 raise last_error
3092 self.report_warning(last_error)
3093 return prs, player_url
3094
3095 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3096 itags, stream_ids = {}, []
3097 itag_qualities, res_qualities = {}, {}
3098 q = qualities([
3099 # Normally tiny is the smallest video-only formats. But
3100 # audio-only formats with unknown quality may get tagged as tiny
3101 'tiny',
3102 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3103 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3104 ])
3105 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3106
3107 for fmt in streaming_formats:
3108 if fmt.get('targetDurationSec'):
3109 continue
3110
3111 itag = str_or_none(fmt.get('itag'))
3112 audio_track = fmt.get('audioTrack') or {}
3113 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3114 if stream_id in stream_ids:
3115 continue
3116
3117 quality = fmt.get('quality')
3118 height = int_or_none(fmt.get('height'))
3119 if quality == 'tiny' or not quality:
3120 quality = fmt.get('audioQuality', '').lower() or quality
3121 # The 3gp format (17) in android client has a quality of "small",
3122 # but is actually worse than other formats
3123 if itag == '17':
3124 quality = 'tiny'
3125 if quality:
3126 if itag:
3127 itag_qualities[itag] = quality
3128 if height:
3129 res_qualities[height] = quality
3130 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3131 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3132 # number of fragment that would subsequently requested with (`&sq=N`)
3133 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3134 continue
3135
3136 fmt_url = fmt.get('url')
3137 if not fmt_url:
3138 sc = compat_parse_qs(fmt.get('signatureCipher'))
3139 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3140 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3141 if not (sc and fmt_url and encrypted_sig):
3142 continue
3143 if not player_url:
3144 continue
3145 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
3146 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
3147 fmt_url += '&' + sp + '=' + signature
3148
3149 query = parse_qs(fmt_url)
3150 throttled = False
3151 if query.get('n'):
3152 try:
3153 fmt_url = update_url_query(fmt_url, {
3154 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3155 except ExtractorError as e:
3156 self.report_warning(
3157 'nsig extraction failed: You may experience throttling for some formats\n'
3158 f'n = {query["n"][0]} ; player = {player_url}\n{e}', only_once=True)
3159 throttled = True
3160
3161 if itag:
3162 itags[itag] = 'https'
3163 stream_ids.append(stream_id)
3164
3165 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3166 language_preference = (
3167 10 if audio_track.get('audioIsDefault') and 10
3168 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3169 else -1)
3170 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3171 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3172 # Make sure to avoid false positives with small duration differences.
3173 # Eg: __2ABJjxzNo, ySuUZEjARPY
3174 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3175 if is_damaged:
3176 self.report_warning(
3177 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3178 dct = {
3179 'asr': int_or_none(fmt.get('audioSampleRate')),
3180 'filesize': int_or_none(fmt.get('contentLength')),
3181 'format_id': itag,
3182 'format_note': join_nonempty(
3183 '%s%s' % (audio_track.get('displayName') or '',
3184 ' (default)' if language_preference > 0 else ''),
3185 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3186 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3187 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3188 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3189 'fps': int_or_none(fmt.get('fps')) or None,
3190 'height': height,
3191 'quality': q(quality),
3192 'has_drm': bool(fmt.get('drmFamilies')),
3193 'tbr': tbr,
3194 'url': fmt_url,
3195 'width': int_or_none(fmt.get('width')),
3196 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3197 'desc' if language_preference < -1 else ''),
3198 'language_preference': language_preference,
3199 # Strictly de-prioritize damaged and 3gp formats
3200 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3201 }
3202 mime_mobj = re.match(
3203 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3204 if mime_mobj:
3205 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3206 dct.update(parse_codecs(mime_mobj.group(2)))
3207 no_audio = dct.get('acodec') == 'none'
3208 no_video = dct.get('vcodec') == 'none'
3209 if no_audio:
3210 dct['vbr'] = tbr
3211 if no_video:
3212 dct['abr'] = tbr
3213 if no_audio or no_video:
3214 dct['downloader_options'] = {
3215 # Youtube throttles chunks >~10M
3216 'http_chunk_size': 10485760,
3217 }
3218 if dct.get('ext'):
3219 dct['container'] = dct['ext'] + '_dash'
3220 yield dct
3221
3222 live_from_start = is_live and self.get_param('live_from_start')
3223 skip_manifests = self._configuration_arg('skip')
3224 if not self.get_param('youtube_include_hls_manifest', True):
3225 skip_manifests.append('hls')
3226 if not self.get_param('youtube_include_dash_manifest', True):
3227 skip_manifests.append('dash')
3228 get_dash = 'dash' not in skip_manifests and (
3229 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3230 get_hls = not live_from_start and 'hls' not in skip_manifests
3231
3232 def process_manifest_format(f, proto, itag):
3233 if itag in itags:
3234 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3235 return False
3236 itag = f'{itag}-{proto}'
3237 if itag:
3238 f['format_id'] = itag
3239 itags[itag] = proto
3240
3241 f['quality'] = next((
3242 q(qdict[val])
3243 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3244 if val in qdict), -1)
3245 return True
3246
3247 for sd in streaming_data:
3248 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3249 if hls_manifest_url:
3250 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3251 if process_manifest_format(f, 'hls', self._search_regex(
3252 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3253 yield f
3254
3255 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3256 if dash_manifest_url:
3257 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3258 if process_manifest_format(f, 'dash', f['format_id']):
3259 f['filesize'] = int_or_none(self._search_regex(
3260 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3261 if live_from_start:
3262 f['is_from_start'] = True
3263
3264 yield f
3265
3266 def _extract_storyboard(self, player_responses, duration):
3267 spec = get_first(
3268 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3269 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3270 if not base_url:
3271 return
3272 L = len(spec) - 1
3273 for i, args in enumerate(spec):
3274 args = args.split('#')
3275 counts = list(map(int_or_none, args[:5]))
3276 if len(args) != 8 or not all(counts):
3277 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3278 continue
3279 width, height, frame_count, cols, rows = counts
3280 N, sigh = args[6:]
3281
3282 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3283 fragment_count = frame_count / (cols * rows)
3284 fragment_duration = duration / fragment_count
3285 yield {
3286 'format_id': f'sb{i}',
3287 'format_note': 'storyboard',
3288 'ext': 'mhtml',
3289 'protocol': 'mhtml',
3290 'acodec': 'none',
3291 'vcodec': 'none',
3292 'url': url,
3293 'width': width,
3294 'height': height,
3295 'fragments': [{
3296 'url': url.replace('$M', str(j)),
3297 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3298 } for j in range(math.ceil(fragment_count))],
3299 }
3300
3301 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3302 webpage = None
3303 if 'webpage' not in self._configuration_arg('player_skip'):
3304 webpage = self._download_webpage(
3305 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3306
3307 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3308
3309 player_responses, player_url = self._extract_player_responses(
3310 self._get_requested_clients(url, smuggled_data),
3311 video_id, webpage, master_ytcfg)
3312
3313 return webpage, master_ytcfg, player_responses, player_url
3314
3315 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3316 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3317 is_live = get_first(video_details, 'isLive')
3318 if is_live is None:
3319 is_live = get_first(live_broadcast_details, 'isLiveNow')
3320
3321 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3322 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3323
3324 return live_broadcast_details, is_live, streaming_data, formats
3325
3326 def _real_extract(self, url):
3327 url, smuggled_data = unsmuggle_url(url, {})
3328 video_id = self._match_id(url)
3329
3330 base_url = self.http_scheme() + '//www.youtube.com/'
3331 webpage_url = base_url + 'watch?v=' + video_id
3332
3333 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3334
3335 playability_statuses = traverse_obj(
3336 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3337
3338 trailer_video_id = get_first(
3339 playability_statuses,
3340 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3341 expected_type=str)
3342 if trailer_video_id:
3343 return self.url_result(
3344 trailer_video_id, self.ie_key(), trailer_video_id)
3345
3346 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3347 if webpage else (lambda x: None))
3348
3349 video_details = traverse_obj(
3350 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3351 microformats = traverse_obj(
3352 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3353 expected_type=dict, default=[])
3354 video_title = (
3355 get_first(video_details, 'title')
3356 or self._get_text(microformats, (..., 'title'))
3357 or search_meta(['og:title', 'twitter:title', 'title']))
3358 video_description = get_first(video_details, 'shortDescription')
3359
3360 multifeed_metadata_list = get_first(
3361 player_responses,
3362 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3363 expected_type=str)
3364 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3365 if self.get_param('noplaylist'):
3366 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3367 else:
3368 entries = []
3369 feed_ids = []
3370 for feed in multifeed_metadata_list.split(','):
3371 # Unquote should take place before split on comma (,) since textual
3372 # fields may contain comma as well (see
3373 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3374 feed_data = compat_parse_qs(
3375 compat_urllib_parse_unquote_plus(feed))
3376
3377 def feed_entry(name):
3378 return try_get(
3379 feed_data, lambda x: x[name][0], compat_str)
3380
3381 feed_id = feed_entry('id')
3382 if not feed_id:
3383 continue
3384 feed_title = feed_entry('title')
3385 title = video_title
3386 if feed_title:
3387 title += ' (%s)' % feed_title
3388 entries.append({
3389 '_type': 'url_transparent',
3390 'ie_key': 'Youtube',
3391 'url': smuggle_url(
3392 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3393 {'force_singlefeed': True}),
3394 'title': title,
3395 })
3396 feed_ids.append(feed_id)
3397 self.to_screen(
3398 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3399 % (', '.join(feed_ids), video_id))
3400 return self.playlist_result(
3401 entries, video_id, video_title, video_description)
3402
3403 duration = int_or_none(
3404 get_first(video_details, 'lengthSeconds')
3405 or get_first(microformats, 'lengthSeconds')
3406 or parse_duration(search_meta('duration'))) or None
3407
3408 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3409 video_id, microformats, video_details, player_responses, player_url, duration)
3410
3411 if not formats:
3412 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3413 self.report_drm(video_id)
3414 pemr = get_first(
3415 playability_statuses,
3416 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3417 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3418 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3419 if subreason:
3420 if subreason == 'The uploader has not made this video available in your country.':
3421 countries = get_first(microformats, 'availableCountries')
3422 if not countries:
3423 regions_allowed = search_meta('regionsAllowed')
3424 countries = regions_allowed.split(',') if regions_allowed else None
3425 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3426 reason += f'. {subreason}'
3427 if reason:
3428 self.raise_no_formats(reason, expected=True)
3429
3430 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3431 if not keywords and webpage:
3432 keywords = [
3433 unescapeHTML(m.group('content'))
3434 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3435 for keyword in keywords:
3436 if keyword.startswith('yt:stretch='):
3437 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3438 if mobj:
3439 # NB: float is intentional for forcing float division
3440 w, h = (float(v) for v in mobj.groups())
3441 if w > 0 and h > 0:
3442 ratio = w / h
3443 for f in formats:
3444 if f.get('vcodec') != 'none':
3445 f['stretched_ratio'] = ratio
3446 break
3447 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3448 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3449 if thumbnail_url:
3450 thumbnails.append({
3451 'url': thumbnail_url,
3452 })
3453 original_thumbnails = thumbnails.copy()
3454
3455 # The best resolution thumbnails sometimes does not appear in the webpage
3456 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3457 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3458 thumbnail_names = [
3459 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3460 # in resolution, these are not the custom thumbnail. So de-prioritize them
3461 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3462 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3463 ]
3464 n_thumbnail_names = len(thumbnail_names)
3465 thumbnails.extend({
3466 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3467 video_id=video_id, name=name, ext=ext,
3468 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3469 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3470 for thumb in thumbnails:
3471 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3472 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3473 self._remove_duplicate_formats(thumbnails)
3474 self._downloader._sort_thumbnails(original_thumbnails)
3475
3476 category = get_first(microformats, 'category') or search_meta('genre')
3477 channel_id = str_or_none(
3478 get_first(video_details, 'channelId')
3479 or get_first(microformats, 'externalChannelId')
3480 or search_meta('channelId'))
3481 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3482
3483 live_content = get_first(video_details, 'isLiveContent')
3484 is_upcoming = get_first(video_details, 'isUpcoming')
3485 if is_live is None:
3486 if is_upcoming or live_content is False:
3487 is_live = False
3488 if is_upcoming is None and (live_content or is_live):
3489 is_upcoming = False
3490 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3491 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3492 if not duration and live_end_time and live_start_time:
3493 duration = live_end_time - live_start_time
3494
3495 if is_live and self.get_param('live_from_start'):
3496 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3497
3498 formats.extend(self._extract_storyboard(player_responses, duration))
3499
3500 # Source is given priority since formats that throttle are given lower source_preference
3501 # When throttling issue is fully fixed, remove this
3502 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3503
3504 info = {
3505 'id': video_id,
3506 'title': video_title,
3507 'formats': formats,
3508 'thumbnails': thumbnails,
3509 # The best thumbnail that we are sure exists. Prevents unnecessary
3510 # URL checking if user don't care about getting the best possible thumbnail
3511 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3512 'description': video_description,
3513 'uploader': get_first(video_details, 'author'),
3514 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3515 'uploader_url': owner_profile_url,
3516 'channel_id': channel_id,
3517 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3518 'duration': duration,
3519 'view_count': int_or_none(
3520 get_first((video_details, microformats), (..., 'viewCount'))
3521 or search_meta('interactionCount')),
3522 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3523 'age_limit': 18 if (
3524 get_first(microformats, 'isFamilySafe') is False
3525 or search_meta('isFamilyFriendly') == 'false'
3526 or search_meta('og:restrictions:age') == '18+') else 0,
3527 'webpage_url': webpage_url,
3528 'categories': [category] if category else None,
3529 'tags': keywords,
3530 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3531 'is_live': is_live,
3532 'was_live': (False if is_live or is_upcoming or live_content is False
3533 else None if is_live is None or is_upcoming is None
3534 else live_content),
3535 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3536 'release_timestamp': live_start_time,
3537 }
3538
3539 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3540 if pctr:
3541 def get_lang_code(track):
3542 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3543 or track.get('languageCode'))
3544
3545 # Converted into dicts to remove duplicates
3546 captions = {
3547 get_lang_code(sub): sub
3548 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3549 translation_languages = {
3550 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3551 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3552
3553 def process_language(container, base_url, lang_code, sub_name, query):
3554 lang_subs = container.setdefault(lang_code, [])
3555 for fmt in self._SUBTITLE_FORMATS:
3556 query.update({
3557 'fmt': fmt,
3558 })
3559 lang_subs.append({
3560 'ext': fmt,
3561 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3562 'name': sub_name,
3563 })
3564
3565 subtitles, automatic_captions = {}, {}
3566 for lang_code, caption_track in captions.items():
3567 base_url = caption_track.get('baseUrl')
3568 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3569 if not base_url:
3570 continue
3571 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3572 if caption_track.get('kind') != 'asr':
3573 if not lang_code:
3574 continue
3575 process_language(
3576 subtitles, base_url, lang_code, lang_name, {})
3577 if not caption_track.get('isTranslatable'):
3578 continue
3579 for trans_code, trans_name in translation_languages.items():
3580 if not trans_code:
3581 continue
3582 orig_trans_code = trans_code
3583 if caption_track.get('kind') != 'asr':
3584 if 'translated_subs' in self._configuration_arg('skip'):
3585 continue
3586 trans_code += f'-{lang_code}'
3587 trans_name += format_field(lang_name, template=' from %s')
3588 # Add an "-orig" label to the original language so that it can be distinguished.
3589 # The subs are returned without "-orig" as well for compatibility
3590 if lang_code == f'a-{orig_trans_code}':
3591 process_language(
3592 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3593 # Setting tlang=lang returns damaged subtitles.
3594 process_language(automatic_captions, base_url, trans_code, trans_name,
3595 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3596 info['automatic_captions'] = automatic_captions
3597 info['subtitles'] = subtitles
3598
3599 parsed_url = compat_urllib_parse_urlparse(url)
3600 for component in [parsed_url.fragment, parsed_url.query]:
3601 query = compat_parse_qs(component)
3602 for k, v in query.items():
3603 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3604 d_k += '_time'
3605 if d_k not in info and k in s_ks:
3606 info[d_k] = parse_duration(query[k][0])
3607
3608 # Youtube Music Auto-generated description
3609 if video_description:
3610 mobj = re.search(r'(?s)(?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?.+\nAuto-generated by YouTube\.\s*$', video_description)
3611 if mobj:
3612 release_year = mobj.group('release_year')
3613 release_date = mobj.group('release_date')
3614 if release_date:
3615 release_date = release_date.replace('-', '')
3616 if not release_year:
3617 release_year = release_date[:4]
3618 info.update({
3619 'album': mobj.group('album'.strip()),
3620 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3621 'track': mobj.group('track').strip(),
3622 'release_date': release_date,
3623 'release_year': int_or_none(release_year),
3624 })
3625
3626 initial_data = None
3627 if webpage:
3628 initial_data = self._extract_yt_initial_variable(
3629 webpage, self._YT_INITIAL_DATA_RE, video_id,
3630 'yt initial data')
3631 if not initial_data:
3632 query = {'videoId': video_id}
3633 query.update(self._get_checkok_params())
3634 initial_data = self._extract_response(
3635 item_id=video_id, ep='next', fatal=False,
3636 ytcfg=master_ytcfg, query=query,
3637 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3638 note='Downloading initial data API JSON')
3639
3640 try: # This will error if there is no livechat
3641 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3642 except (KeyError, IndexError, TypeError):
3643 pass
3644 else:
3645 info.setdefault('subtitles', {})['live_chat'] = [{
3646 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3647 'video_id': video_id,
3648 'ext': 'json',
3649 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3650 }]
3651
3652 if initial_data:
3653 info['chapters'] = (
3654 self._extract_chapters_from_json(initial_data, duration)
3655 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3656 or None)
3657
3658 contents = traverse_obj(
3659 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3660 expected_type=list, default=[])
3661
3662 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3663 if vpir:
3664 stl = vpir.get('superTitleLink')
3665 if stl:
3666 stl = self._get_text(stl)
3667 if try_get(
3668 vpir,
3669 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3670 info['location'] = stl
3671 else:
3672 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3673 if mobj:
3674 info.update({
3675 'series': mobj.group(1),
3676 'season_number': int(mobj.group(2)),
3677 'episode_number': int(mobj.group(3)),
3678 })
3679 for tlb in (try_get(
3680 vpir,
3681 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3682 list) or []):
3683 tbr = tlb.get('toggleButtonRenderer') or {}
3684 for getter, regex in [(
3685 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3686 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3687 lambda x: x['accessibility'],
3688 lambda x: x['accessibilityData']['accessibilityData'],
3689 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3690 label = (try_get(tbr, getter, dict) or {}).get('label')
3691 if label:
3692 mobj = re.match(regex, label)
3693 if mobj:
3694 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3695 break
3696 sbr_tooltip = try_get(
3697 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3698 if sbr_tooltip:
3699 like_count, dislike_count = sbr_tooltip.split(' / ')
3700 info.update({
3701 'like_count': str_to_int(like_count),
3702 'dislike_count': str_to_int(dislike_count),
3703 })
3704 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3705 if vsir:
3706 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3707 info.update({
3708 'channel': self._get_text(vor, 'title'),
3709 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3710
3711 rows = try_get(
3712 vsir,
3713 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3714 list) or []
3715 multiple_songs = False
3716 for row in rows:
3717 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3718 multiple_songs = True
3719 break
3720 for row in rows:
3721 mrr = row.get('metadataRowRenderer') or {}
3722 mrr_title = mrr.get('title')
3723 if not mrr_title:
3724 continue
3725 mrr_title = self._get_text(mrr, 'title')
3726 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3727 if mrr_title == 'License':
3728 info['license'] = mrr_contents_text
3729 elif not multiple_songs:
3730 if mrr_title == 'Album':
3731 info['album'] = mrr_contents_text
3732 elif mrr_title == 'Artist':
3733 info['artist'] = mrr_contents_text
3734 elif mrr_title == 'Song':
3735 info['track'] = mrr_contents_text
3736
3737 fallbacks = {
3738 'channel': 'uploader',
3739 'channel_id': 'uploader_id',
3740 'channel_url': 'uploader_url',
3741 }
3742
3743 # The upload date for scheduled, live and past live streams / premieres in microformats
3744 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3745 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3746 upload_date = (
3747 unified_strdate(get_first(microformats, 'uploadDate'))
3748 or unified_strdate(search_meta('uploadDate')))
3749 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3750 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3751 info['upload_date'] = upload_date
3752
3753 for to, frm in fallbacks.items():
3754 if not info.get(to):
3755 info[to] = info.get(frm)
3756
3757 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3758 v = info.get(s_k)
3759 if v:
3760 info[d_k] = v
3761
3762 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3763 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3764 is_membersonly = None
3765 is_premium = None
3766 if initial_data and is_private is not None:
3767 is_membersonly = False
3768 is_premium = False
3769 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3770 badge_labels = set()
3771 for content in contents:
3772 if not isinstance(content, dict):
3773 continue
3774 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3775 for badge_label in badge_labels:
3776 if badge_label.lower() == 'members only':
3777 is_membersonly = True
3778 elif badge_label.lower() == 'premium':
3779 is_premium = True
3780 elif badge_label.lower() == 'unlisted':
3781 is_unlisted = True
3782
3783 info['availability'] = self._availability(
3784 is_private=is_private,
3785 needs_premium=is_premium,
3786 needs_subscription=is_membersonly,
3787 needs_auth=info['age_limit'] >= 18,
3788 is_unlisted=None if is_private is None else is_unlisted)
3789
3790 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3791
3792 self.mark_watched(video_id, player_responses)
3793
3794 return info
3795
3796
3797 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3798
3799 @staticmethod
3800 def passthrough_smuggled_data(func):
3801 def _smuggle(entries, smuggled_data):
3802 for entry in entries:
3803 # TODO: Convert URL to music.youtube instead.
3804 # Do we need to passthrough any other smuggled_data?
3805 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3806 yield entry
3807
3808 @functools.wraps(func)
3809 def wrapper(self, url):
3810 url, smuggled_data = unsmuggle_url(url, {})
3811 if self.is_music_url(url):
3812 smuggled_data['is_music_url'] = True
3813 info_dict = func(self, url, smuggled_data)
3814 if smuggled_data and info_dict.get('entries'):
3815 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3816 return info_dict
3817 return wrapper
3818
3819 def _extract_channel_id(self, webpage):
3820 channel_id = self._html_search_meta(
3821 'channelId', webpage, 'channel id', default=None)
3822 if channel_id:
3823 return channel_id
3824 channel_url = self._html_search_meta(
3825 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3826 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3827 'twitter:app:url:googleplay'), webpage, 'channel url')
3828 return self._search_regex(
3829 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3830 channel_url, 'channel id')
3831
3832 @staticmethod
3833 def _extract_basic_item_renderer(item):
3834 # Modified from _extract_grid_item_renderer
3835 known_basic_renderers = (
3836 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3837 )
3838 for key, renderer in item.items():
3839 if not isinstance(renderer, dict):
3840 continue
3841 elif key in known_basic_renderers:
3842 return renderer
3843 elif key.startswith('grid') and key.endswith('Renderer'):
3844 return renderer
3845
3846 def _grid_entries(self, grid_renderer):
3847 for item in grid_renderer['items']:
3848 if not isinstance(item, dict):
3849 continue
3850 renderer = self._extract_basic_item_renderer(item)
3851 if not isinstance(renderer, dict):
3852 continue
3853 title = self._get_text(renderer, 'title')
3854
3855 # playlist
3856 playlist_id = renderer.get('playlistId')
3857 if playlist_id:
3858 yield self.url_result(
3859 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3860 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3861 video_title=title)
3862 continue
3863 # video
3864 video_id = renderer.get('videoId')
3865 if video_id:
3866 yield self._extract_video(renderer)
3867 continue
3868 # channel
3869 channel_id = renderer.get('channelId')
3870 if channel_id:
3871 yield self.url_result(
3872 'https://www.youtube.com/channel/%s' % channel_id,
3873 ie=YoutubeTabIE.ie_key(), video_title=title)
3874 continue
3875 # generic endpoint URL support
3876 ep_url = urljoin('https://www.youtube.com/', try_get(
3877 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3878 compat_str))
3879 if ep_url:
3880 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3881 if ie.suitable(ep_url):
3882 yield self.url_result(
3883 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3884 break
3885
3886 def _music_reponsive_list_entry(self, renderer):
3887 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3888 if video_id:
3889 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3890 ie=YoutubeIE.ie_key(), video_id=video_id)
3891 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3892 if playlist_id:
3893 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3894 if video_id:
3895 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3896 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3897 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3898 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3899 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3900 if browse_id:
3901 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3902 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3903
3904 def _shelf_entries_from_content(self, shelf_renderer):
3905 content = shelf_renderer.get('content')
3906 if not isinstance(content, dict):
3907 return
3908 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3909 if renderer:
3910 # TODO: add support for nested playlists so each shelf is processed
3911 # as separate playlist
3912 # TODO: this includes only first N items
3913 yield from self._grid_entries(renderer)
3914 renderer = content.get('horizontalListRenderer')
3915 if renderer:
3916 # TODO
3917 pass
3918
3919 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3920 ep = try_get(
3921 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3922 compat_str)
3923 shelf_url = urljoin('https://www.youtube.com', ep)
3924 if shelf_url:
3925 # Skipping links to another channels, note that checking for
3926 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3927 # will not work
3928 if skip_channels and '/channels?' in shelf_url:
3929 return
3930 title = self._get_text(shelf_renderer, 'title')
3931 yield self.url_result(shelf_url, video_title=title)
3932 # Shelf may not contain shelf URL, fallback to extraction from content
3933 yield from self._shelf_entries_from_content(shelf_renderer)
3934
3935 def _playlist_entries(self, video_list_renderer):
3936 for content in video_list_renderer['contents']:
3937 if not isinstance(content, dict):
3938 continue
3939 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3940 if not isinstance(renderer, dict):
3941 continue
3942 video_id = renderer.get('videoId')
3943 if not video_id:
3944 continue
3945 yield self._extract_video(renderer)
3946
3947 def _rich_entries(self, rich_grid_renderer):
3948 renderer = try_get(
3949 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3950 video_id = renderer.get('videoId')
3951 if not video_id:
3952 return
3953 yield self._extract_video(renderer)
3954
3955 def _video_entry(self, video_renderer):
3956 video_id = video_renderer.get('videoId')
3957 if video_id:
3958 return self._extract_video(video_renderer)
3959
3960 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3961 url = urljoin('https://youtube.com', traverse_obj(
3962 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3963 if url:
3964 return self.url_result(
3965 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3966
3967 def _post_thread_entries(self, post_thread_renderer):
3968 post_renderer = try_get(
3969 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3970 if not post_renderer:
3971 return
3972 # video attachment
3973 video_renderer = try_get(
3974 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3975 video_id = video_renderer.get('videoId')
3976 if video_id:
3977 entry = self._extract_video(video_renderer)
3978 if entry:
3979 yield entry
3980 # playlist attachment
3981 playlist_id = try_get(
3982 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3983 if playlist_id:
3984 yield self.url_result(
3985 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3986 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3987 # inline video links
3988 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
3989 for run in runs:
3990 if not isinstance(run, dict):
3991 continue
3992 ep_url = try_get(
3993 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
3994 if not ep_url:
3995 continue
3996 if not YoutubeIE.suitable(ep_url):
3997 continue
3998 ep_video_id = YoutubeIE._match_id(ep_url)
3999 if video_id == ep_video_id:
4000 continue
4001 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4002
4003 def _post_thread_continuation_entries(self, post_thread_continuation):
4004 contents = post_thread_continuation.get('contents')
4005 if not isinstance(contents, list):
4006 return
4007 for content in contents:
4008 renderer = content.get('backstagePostThreadRenderer')
4009 if not isinstance(renderer, dict):
4010 continue
4011 yield from self._post_thread_entries(renderer)
4012
4013 r''' # unused
4014 def _rich_grid_entries(self, contents):
4015 for content in contents:
4016 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4017 if video_renderer:
4018 entry = self._video_entry(video_renderer)
4019 if entry:
4020 yield entry
4021 '''
4022
4023 def _extract_entries(self, parent_renderer, continuation_list):
4024 # continuation_list is modified in-place with continuation_list = [continuation_token]
4025 continuation_list[:] = [None]
4026 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4027 for content in contents:
4028 if not isinstance(content, dict):
4029 continue
4030 is_renderer = traverse_obj(
4031 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4032 expected_type=dict)
4033 if not is_renderer:
4034 renderer = content.get('richItemRenderer')
4035 if renderer:
4036 for entry in self._rich_entries(renderer):
4037 yield entry
4038 continuation_list[0] = self._extract_continuation(parent_renderer)
4039 continue
4040 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4041 for isr_content in isr_contents:
4042 if not isinstance(isr_content, dict):
4043 continue
4044
4045 known_renderers = {
4046 'playlistVideoListRenderer': self._playlist_entries,
4047 'gridRenderer': self._grid_entries,
4048 'reelShelfRenderer': self._grid_entries,
4049 'shelfRenderer': self._shelf_entries,
4050 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4051 'backstagePostThreadRenderer': self._post_thread_entries,
4052 'videoRenderer': lambda x: [self._video_entry(x)],
4053 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4054 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4055 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4056 }
4057 for key, renderer in isr_content.items():
4058 if key not in known_renderers:
4059 continue
4060 for entry in known_renderers[key](renderer):
4061 if entry:
4062 yield entry
4063 continuation_list[0] = self._extract_continuation(renderer)
4064 break
4065
4066 if not continuation_list[0]:
4067 continuation_list[0] = self._extract_continuation(is_renderer)
4068
4069 if not continuation_list[0]:
4070 continuation_list[0] = self._extract_continuation(parent_renderer)
4071
4072 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4073 continuation_list = [None]
4074 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4075 tab_content = try_get(tab, lambda x: x['content'], dict)
4076 if not tab_content:
4077 return
4078 parent_renderer = (
4079 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4080 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4081 yield from extract_entries(parent_renderer)
4082 continuation = continuation_list[0]
4083
4084 for page_num in itertools.count(1):
4085 if not continuation:
4086 break
4087 headers = self.generate_api_headers(
4088 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4089 response = self._extract_response(
4090 item_id=f'{item_id} page {page_num}',
4091 query=continuation, headers=headers, ytcfg=ytcfg,
4092 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4093
4094 if not response:
4095 break
4096 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4097 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4098 visitor_data = self._extract_visitor_data(response) or visitor_data
4099
4100 known_continuation_renderers = {
4101 'playlistVideoListContinuation': self._playlist_entries,
4102 'gridContinuation': self._grid_entries,
4103 'itemSectionContinuation': self._post_thread_continuation_entries,
4104 'sectionListContinuation': extract_entries, # for feeds
4105 }
4106 continuation_contents = try_get(
4107 response, lambda x: x['continuationContents'], dict) or {}
4108 continuation_renderer = None
4109 for key, value in continuation_contents.items():
4110 if key not in known_continuation_renderers:
4111 continue
4112 continuation_renderer = value
4113 continuation_list = [None]
4114 yield from known_continuation_renderers[key](continuation_renderer)
4115 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4116 break
4117 if continuation_renderer:
4118 continue
4119
4120 known_renderers = {
4121 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4122 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4123 'gridVideoRenderer': (self._grid_entries, 'items'),
4124 'gridChannelRenderer': (self._grid_entries, 'items'),
4125 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4126 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4127 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4128 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4129 }
4130 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4131 continuation_items = try_get(
4132 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4133 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4134 video_items_renderer = None
4135 for key, value in continuation_item.items():
4136 if key not in known_renderers:
4137 continue
4138 video_items_renderer = {known_renderers[key][1]: continuation_items}
4139 continuation_list = [None]
4140 yield from known_renderers[key][0](video_items_renderer)
4141 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4142 break
4143 if video_items_renderer:
4144 continue
4145 break
4146
4147 @staticmethod
4148 def _extract_selected_tab(tabs, fatal=True):
4149 for tab in tabs:
4150 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4151 if renderer.get('selected') is True:
4152 return renderer
4153 else:
4154 if fatal:
4155 raise ExtractorError('Unable to find selected tab')
4156
4157 def _extract_uploader(self, data):
4158 uploader = {}
4159 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4160 owner = try_get(
4161 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4162 if owner:
4163 owner_text = owner.get('text')
4164 uploader['uploader'] = self._search_regex(
4165 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4166 uploader['uploader_id'] = try_get(
4167 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4168 uploader['uploader_url'] = urljoin(
4169 'https://www.youtube.com/',
4170 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4171 return {k: v for k, v in uploader.items() if v is not None}
4172
4173 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4174 playlist_id = title = description = channel_url = channel_name = channel_id = None
4175 tags = []
4176
4177 selected_tab = self._extract_selected_tab(tabs)
4178 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4179 renderer = try_get(
4180 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4181 if renderer:
4182 channel_name = renderer.get('title')
4183 channel_url = renderer.get('channelUrl')
4184 channel_id = renderer.get('externalId')
4185 else:
4186 renderer = try_get(
4187 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4188
4189 if renderer:
4190 title = renderer.get('title')
4191 description = renderer.get('description', '')
4192 playlist_id = channel_id
4193 tags = renderer.get('keywords', '').split()
4194
4195 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4196 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4197 def _get_uncropped(url):
4198 return url_or_none((url or '').split('=')[0] + '=s0')
4199
4200 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4201 if avatar_thumbnails:
4202 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4203 if uncropped_avatar:
4204 avatar_thumbnails.append({
4205 'url': uncropped_avatar,
4206 'id': 'avatar_uncropped',
4207 'preference': 1
4208 })
4209
4210 channel_banners = self._extract_thumbnails(
4211 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4212 for banner in channel_banners:
4213 banner['preference'] = -10
4214
4215 if channel_banners:
4216 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4217 if uncropped_banner:
4218 channel_banners.append({
4219 'url': uncropped_banner,
4220 'id': 'banner_uncropped',
4221 'preference': -5
4222 })
4223
4224 primary_thumbnails = self._extract_thumbnails(
4225 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4226
4227 if playlist_id is None:
4228 playlist_id = item_id
4229
4230 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4231 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4232 if title is None:
4233 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4234 title += format_field(selected_tab, 'title', ' - %s')
4235 title += format_field(selected_tab, 'expandedText', ' - %s')
4236
4237 metadata = {
4238 'playlist_id': playlist_id,
4239 'playlist_title': title,
4240 'playlist_description': description,
4241 'uploader': channel_name,
4242 'uploader_id': channel_id,
4243 'uploader_url': channel_url,
4244 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4245 'tags': tags,
4246 'view_count': self._get_count(playlist_stats, 1),
4247 'availability': self._extract_availability(data),
4248 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4249 'playlist_count': self._get_count(playlist_stats, 0),
4250 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4251 }
4252 if not channel_id:
4253 metadata.update(self._extract_uploader(data))
4254 metadata.update({
4255 'channel': metadata['uploader'],
4256 'channel_id': metadata['uploader_id'],
4257 'channel_url': metadata['uploader_url']})
4258 return self.playlist_result(
4259 self._entries(
4260 selected_tab, playlist_id, ytcfg,
4261 self._extract_account_syncid(ytcfg, data),
4262 self._extract_visitor_data(data, ytcfg)),
4263 **metadata)
4264
4265 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4266 first_id = last_id = response = None
4267 for page_num in itertools.count(1):
4268 videos = list(self._playlist_entries(playlist))
4269 if not videos:
4270 return
4271 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4272 if start >= len(videos):
4273 return
4274 for video in videos[start:]:
4275 yield video
4276 first_id = first_id or videos[0]['id']
4277 last_id = videos[-1]['id']
4278 watch_endpoint = try_get(
4279 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4280 headers = self.generate_api_headers(
4281 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4282 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4283 query = {
4284 'playlistId': playlist_id,
4285 'videoId': watch_endpoint.get('videoId') or last_id,
4286 'index': watch_endpoint.get('index') or len(videos),
4287 'params': watch_endpoint.get('params') or 'OAE%3D'
4288 }
4289 response = self._extract_response(
4290 item_id='%s page %d' % (playlist_id, page_num),
4291 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4292 check_get_keys='contents'
4293 )
4294 playlist = try_get(
4295 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4296
4297 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4298 title = playlist.get('title') or try_get(
4299 data, lambda x: x['titleText']['simpleText'], compat_str)
4300 playlist_id = playlist.get('playlistId') or item_id
4301
4302 # Delegating everything except mix playlists to regular tab-based playlist URL
4303 playlist_url = urljoin(url, try_get(
4304 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4305 compat_str))
4306
4307 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4308 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4309 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4310
4311 if playlist_url and playlist_url != url and not is_known_unviewable:
4312 return self.url_result(
4313 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4314 video_title=title)
4315
4316 return self.playlist_result(
4317 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4318 playlist_id=playlist_id, playlist_title=title)
4319
4320 def _extract_availability(self, data):
4321 """
4322 Gets the availability of a given playlist/tab.
4323 Note: Unless YouTube tells us explicitly, we do not assume it is public
4324 @param data: response
4325 """
4326 is_private = is_unlisted = None
4327 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4328 badge_labels = self._extract_badges(renderer)
4329
4330 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4331 privacy_dropdown_entries = try_get(
4332 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4333 for renderer_dict in privacy_dropdown_entries:
4334 is_selected = try_get(
4335 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4336 if not is_selected:
4337 continue
4338 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4339 if label:
4340 badge_labels.add(label.lower())
4341 break
4342
4343 for badge_label in badge_labels:
4344 if badge_label == 'unlisted':
4345 is_unlisted = True
4346 elif badge_label == 'private':
4347 is_private = True
4348 elif badge_label == 'public':
4349 is_unlisted = is_private = False
4350 return self._availability(is_private, False, False, False, is_unlisted)
4351
4352 @staticmethod
4353 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4354 sidebar_renderer = try_get(
4355 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4356 for item in sidebar_renderer:
4357 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4358 if renderer:
4359 return renderer
4360
4361 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4362 """
4363 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4364 """
4365 browse_id = params = None
4366 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4367 if not renderer:
4368 return
4369 menu_renderer = try_get(
4370 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4371 for menu_item in menu_renderer:
4372 if not isinstance(menu_item, dict):
4373 continue
4374 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4375 text = try_get(
4376 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4377 if not text or text.lower() != 'show unavailable videos':
4378 continue
4379 browse_endpoint = try_get(
4380 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4381 browse_id = browse_endpoint.get('browseId')
4382 params = browse_endpoint.get('params')
4383 break
4384
4385 headers = self.generate_api_headers(
4386 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4387 visitor_data=self._extract_visitor_data(data, ytcfg))
4388 query = {
4389 'params': params or 'wgYCCAA=',
4390 'browseId': browse_id or 'VL%s' % item_id
4391 }
4392 return self._extract_response(
4393 item_id=item_id, headers=headers, query=query,
4394 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4395 note='Downloading API JSON with unavailable videos')
4396
4397 @property
4398 def skip_webpage(self):
4399 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4400
4401 def _extract_webpage(self, url, item_id, fatal=True):
4402 retries = self.get_param('extractor_retries', 3)
4403 count = -1
4404 webpage = data = last_error = None
4405 while count < retries:
4406 count += 1
4407 # Sometimes youtube returns a webpage with incomplete ytInitialData
4408 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4409 if last_error:
4410 self.report_warning('%s. Retrying ...' % last_error)
4411 try:
4412 webpage = self._download_webpage(
4413 url, item_id,
4414 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4415 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4416 except ExtractorError as e:
4417 if isinstance(e.cause, network_exceptions):
4418 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4419 last_error = error_to_compat_str(e.cause or e.msg)
4420 if count < retries:
4421 continue
4422 if fatal:
4423 raise
4424 self.report_warning(error_to_compat_str(e))
4425 break
4426 else:
4427 try:
4428 self._extract_and_report_alerts(data)
4429 except ExtractorError as e:
4430 if fatal:
4431 raise
4432 self.report_warning(error_to_compat_str(e))
4433 break
4434
4435 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4436 break
4437
4438 last_error = 'Incomplete yt initial data received'
4439 if count >= retries:
4440 if fatal:
4441 raise ExtractorError(last_error)
4442 self.report_warning(last_error)
4443 break
4444
4445 return webpage, data
4446
4447 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4448 """Use if failed to extract ytcfg (and data) from initial webpage"""
4449 if not ytcfg and self.is_authenticated:
4450 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4451 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4452 raise ExtractorError(
4453 f'{msg}. If you are not downloading private content, or '
4454 'your cookies are only for the first account and channel,'
4455 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4456 expected=True)
4457 self.report_warning(msg, only_once=True)
4458
4459 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4460 data = None
4461 if not self.skip_webpage:
4462 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4463 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4464 # Reject webpage data if redirected to home page without explicitly requesting
4465 selected_tab = self._extract_selected_tab(traverse_obj(
4466 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4467 if (url != 'https://www.youtube.com/feed/recommended'
4468 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4469 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4470 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4471 if fatal:
4472 raise ExtractorError(msg, expected=True)
4473 self.report_warning(msg, only_once=True)
4474 if not data:
4475 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4476 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4477 return data, ytcfg
4478
4479 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4480 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4481 resolve_response = self._extract_response(
4482 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4483 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4484 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4485 for ep_key, ep in endpoints.items():
4486 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4487 if params:
4488 return self._extract_response(
4489 item_id=item_id, query=params, ep=ep, headers=headers,
4490 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4491 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4492 err_note = 'Failed to resolve url (does the playlist exist?)'
4493 if fatal:
4494 raise ExtractorError(err_note, expected=True)
4495 self.report_warning(err_note, item_id)
4496
4497 _SEARCH_PARAMS = None
4498
4499 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4500 data = {'query': query}
4501 if params is NO_DEFAULT:
4502 params = self._SEARCH_PARAMS
4503 if params:
4504 data['params'] = params
4505
4506 content_keys = (
4507 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4508 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4509 # ytmusic search
4510 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4511 ('continuationContents', ),
4512 )
4513 display_id = f'query "{query}"'
4514 check_get_keys = tuple({keys[0] for keys in content_keys})
4515 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4516 self._report_playlist_authcheck(ytcfg, fatal=False)
4517
4518 continuation_list = [None]
4519 search = None
4520 for page_num in itertools.count(1):
4521 data.update(continuation_list[0] or {})
4522 headers = self.generate_api_headers(
4523 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4524 search = self._extract_response(
4525 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4526 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4527 slr_contents = traverse_obj(search, *content_keys)
4528 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4529 if not continuation_list[0]:
4530 break
4531
4532
4533 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4534 IE_DESC = 'YouTube Tabs'
4535 _VALID_URL = r'''(?x:
4536 https?://
4537 (?:\w+\.)?
4538 (?:
4539 youtube(?:kids)?\.com|
4540 %(invidious)s
4541 )/
4542 (?:
4543 (?P<channel_type>channel|c|user|browse)/|
4544 (?P<not_channel>
4545 feed/|hashtag/|
4546 (?:playlist|watch)\?.*?\blist=
4547 )|
4548 (?!(?:%(reserved_names)s)\b) # Direct URLs
4549 )
4550 (?P<id>[^/?\#&]+)
4551 )''' % {
4552 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4553 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4554 }
4555 IE_NAME = 'youtube:tab'
4556
4557 _TESTS = [{
4558 'note': 'playlists, multipage',
4559 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4560 'playlist_mincount': 94,
4561 'info_dict': {
4562 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4563 'title': 'Igor Kleiner - Playlists',
4564 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4565 'uploader': 'Igor Kleiner',
4566 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4567 'channel': 'Igor Kleiner',
4568 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4569 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4570 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4571 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4572 'channel_follower_count': int
4573 },
4574 }, {
4575 'note': 'playlists, multipage, different order',
4576 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4577 'playlist_mincount': 94,
4578 'info_dict': {
4579 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4580 'title': 'Igor Kleiner - Playlists',
4581 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4582 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4583 'uploader': 'Igor Kleiner',
4584 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4585 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4586 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4587 'channel': 'Igor Kleiner',
4588 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4589 'channel_follower_count': int
4590 },
4591 }, {
4592 'note': 'playlists, series',
4593 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4594 'playlist_mincount': 5,
4595 'info_dict': {
4596 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4597 'title': '3Blue1Brown - Playlists',
4598 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4599 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4600 'uploader': '3Blue1Brown',
4601 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4602 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4603 'channel': '3Blue1Brown',
4604 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4605 'tags': ['Mathematics'],
4606 'channel_follower_count': int
4607 },
4608 }, {
4609 'note': 'playlists, singlepage',
4610 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4611 'playlist_mincount': 4,
4612 'info_dict': {
4613 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4614 'title': 'ThirstForScience - Playlists',
4615 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4616 'uploader': 'ThirstForScience',
4617 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4618 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4619 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4620 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4621 'tags': 'count:13',
4622 'channel': 'ThirstForScience',
4623 'channel_follower_count': int
4624 }
4625 }, {
4626 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4627 'only_matching': True,
4628 }, {
4629 'note': 'basic, single video playlist',
4630 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4631 'info_dict': {
4632 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4633 'uploader': 'Sergey M.',
4634 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4635 'title': 'youtube-dl public playlist',
4636 'description': '',
4637 'tags': [],
4638 'view_count': int,
4639 'modified_date': '20201130',
4640 'channel': 'Sergey M.',
4641 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4642 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4643 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4644 },
4645 'playlist_count': 1,
4646 }, {
4647 'note': 'empty playlist',
4648 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4649 'info_dict': {
4650 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4651 'uploader': 'Sergey M.',
4652 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4653 'title': 'youtube-dl empty playlist',
4654 'tags': [],
4655 'channel': 'Sergey M.',
4656 'description': '',
4657 'modified_date': '20160902',
4658 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4659 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4660 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4661 },
4662 'playlist_count': 0,
4663 }, {
4664 'note': 'Home tab',
4665 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4666 'info_dict': {
4667 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4668 'title': 'lex will - Home',
4669 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4670 'uploader': 'lex will',
4671 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4672 'channel': 'lex will',
4673 'tags': ['bible', 'history', 'prophesy'],
4674 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4675 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4676 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4677 'channel_follower_count': int
4678 },
4679 'playlist_mincount': 2,
4680 }, {
4681 'note': 'Videos tab',
4682 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4683 'info_dict': {
4684 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4685 'title': 'lex will - Videos',
4686 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4687 'uploader': 'lex will',
4688 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4689 'tags': ['bible', 'history', 'prophesy'],
4690 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4691 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4692 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4693 'channel': 'lex will',
4694 'channel_follower_count': int
4695 },
4696 'playlist_mincount': 975,
4697 }, {
4698 'note': 'Videos tab, sorted by popular',
4699 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4700 'info_dict': {
4701 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4702 'title': 'lex will - Videos',
4703 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4704 'uploader': 'lex will',
4705 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4706 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4707 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4708 'channel': 'lex will',
4709 'tags': ['bible', 'history', 'prophesy'],
4710 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4711 'channel_follower_count': int
4712 },
4713 'playlist_mincount': 199,
4714 }, {
4715 'note': 'Playlists tab',
4716 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4717 'info_dict': {
4718 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4719 'title': 'lex will - Playlists',
4720 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4721 'uploader': 'lex will',
4722 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4723 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4724 'channel': 'lex will',
4725 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4726 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4727 'tags': ['bible', 'history', 'prophesy'],
4728 'channel_follower_count': int
4729 },
4730 'playlist_mincount': 17,
4731 }, {
4732 'note': 'Community tab',
4733 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4734 'info_dict': {
4735 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4736 'title': 'lex will - Community',
4737 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4738 'uploader': 'lex will',
4739 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4740 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'channel': 'lex will',
4742 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4743 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4744 'tags': ['bible', 'history', 'prophesy'],
4745 'channel_follower_count': int
4746 },
4747 'playlist_mincount': 18,
4748 }, {
4749 'note': 'Channels tab',
4750 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4751 'info_dict': {
4752 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4753 'title': 'lex will - Channels',
4754 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4755 'uploader': 'lex will',
4756 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4757 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4758 'channel': 'lex will',
4759 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4760 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4761 'tags': ['bible', 'history', 'prophesy'],
4762 'channel_follower_count': int
4763 },
4764 'playlist_mincount': 12,
4765 }, {
4766 'note': 'Search tab',
4767 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4768 'playlist_mincount': 40,
4769 'info_dict': {
4770 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4771 'title': '3Blue1Brown - Search - linear algebra',
4772 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4773 'uploader': '3Blue1Brown',
4774 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4775 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4776 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4777 'tags': ['Mathematics'],
4778 'channel': '3Blue1Brown',
4779 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4780 'channel_follower_count': int
4781 },
4782 }, {
4783 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4784 'only_matching': True,
4785 }, {
4786 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4787 'only_matching': True,
4788 }, {
4789 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4790 'only_matching': True,
4791 }, {
4792 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4793 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4794 'info_dict': {
4795 'title': '29C3: Not my department',
4796 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4797 'uploader': 'Christiaan008',
4798 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4799 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4800 'tags': [],
4801 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4802 'view_count': int,
4803 'modified_date': '20150605',
4804 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4805 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4806 'channel': 'Christiaan008',
4807 },
4808 'playlist_count': 96,
4809 }, {
4810 'note': 'Large playlist',
4811 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4812 'info_dict': {
4813 'title': 'Uploads from Cauchemar',
4814 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4815 'uploader': 'Cauchemar',
4816 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4817 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4818 'tags': [],
4819 'modified_date': r're:\d{8}',
4820 'channel': 'Cauchemar',
4821 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4822 'view_count': int,
4823 'description': '',
4824 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4825 },
4826 'playlist_mincount': 1123,
4827 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4828 }, {
4829 'note': 'even larger playlist, 8832 videos',
4830 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4831 'only_matching': True,
4832 }, {
4833 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4834 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4835 'info_dict': {
4836 'title': 'Uploads from Interstellar Movie',
4837 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4838 'uploader': 'Interstellar Movie',
4839 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4840 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4841 'tags': [],
4842 'view_count': int,
4843 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4844 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4845 'channel': 'Interstellar Movie',
4846 'description': '',
4847 'modified_date': r're:\d{8}',
4848 },
4849 'playlist_mincount': 21,
4850 }, {
4851 'note': 'Playlist with "show unavailable videos" button',
4852 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4853 'info_dict': {
4854 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4855 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4856 'uploader': 'Phim Siêu Nhân Nhật Bản',
4857 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4858 'view_count': int,
4859 'channel': 'Phim Siêu Nhân Nhật Bản',
4860 'tags': [],
4861 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4862 'description': '',
4863 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4864 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4865 'modified_date': r're:\d{8}',
4866 },
4867 'playlist_mincount': 200,
4868 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4869 }, {
4870 'note': 'Playlist with unavailable videos in page 7',
4871 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4872 'info_dict': {
4873 'title': 'Uploads from BlankTV',
4874 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4875 'uploader': 'BlankTV',
4876 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4877 'channel': 'BlankTV',
4878 'channel_url': 'https://www.youtube.com/c/blanktv',
4879 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4880 'view_count': int,
4881 'tags': [],
4882 'uploader_url': 'https://www.youtube.com/c/blanktv',
4883 'modified_date': r're:\d{8}',
4884 'description': '',
4885 },
4886 'playlist_mincount': 1000,
4887 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4888 }, {
4889 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4890 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4891 'info_dict': {
4892 'title': 'Data Analysis with Dr Mike Pound',
4893 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4894 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4895 'uploader': 'Computerphile',
4896 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4897 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4898 'tags': [],
4899 'view_count': int,
4900 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4901 'channel_url': 'https://www.youtube.com/user/Computerphile',
4902 'channel': 'Computerphile',
4903 },
4904 'playlist_mincount': 11,
4905 }, {
4906 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4907 'only_matching': True,
4908 }, {
4909 'note': 'Playlist URL that does not actually serve a playlist',
4910 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4911 'info_dict': {
4912 'id': 'FqZTN594JQw',
4913 'ext': 'webm',
4914 'title': "Smiley's People 01 detective, Adventure Series, Action",
4915 'uploader': 'STREEM',
4916 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4917 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4918 'upload_date': '20150526',
4919 'license': 'Standard YouTube License',
4920 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4921 'categories': ['People & Blogs'],
4922 'tags': list,
4923 'view_count': int,
4924 'like_count': int,
4925 },
4926 'params': {
4927 'skip_download': True,
4928 },
4929 'skip': 'This video is not available.',
4930 'add_ie': [YoutubeIE.ie_key()],
4931 }, {
4932 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4933 'only_matching': True,
4934 }, {
4935 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4936 'only_matching': True,
4937 }, {
4938 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4939 'info_dict': {
4940 'id': 'GgL890LIznQ', # This will keep changing
4941 'ext': 'mp4',
4942 'title': str,
4943 'uploader': 'Sky News',
4944 'uploader_id': 'skynews',
4945 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4946 'upload_date': r're:\d{8}',
4947 'description': str,
4948 'categories': ['News & Politics'],
4949 'tags': list,
4950 'like_count': int,
4951 'release_timestamp': 1642502819,
4952 'channel': 'Sky News',
4953 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4954 'age_limit': 0,
4955 'view_count': int,
4956 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4957 'playable_in_embed': True,
4958 'release_date': '20220118',
4959 'availability': 'public',
4960 'live_status': 'is_live',
4961 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4962 'channel_follower_count': int
4963 },
4964 'params': {
4965 'skip_download': True,
4966 },
4967 'expected_warnings': ['Ignoring subtitle tracks found in '],
4968 }, {
4969 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4970 'info_dict': {
4971 'id': 'a48o2S1cPoo',
4972 'ext': 'mp4',
4973 'title': 'The Young Turks - Live Main Show',
4974 'uploader': 'The Young Turks',
4975 'uploader_id': 'TheYoungTurks',
4976 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4977 'upload_date': '20150715',
4978 'license': 'Standard YouTube License',
4979 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4980 'categories': ['News & Politics'],
4981 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4982 'like_count': int,
4983 },
4984 'params': {
4985 'skip_download': True,
4986 },
4987 'only_matching': True,
4988 }, {
4989 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
4990 'only_matching': True,
4991 }, {
4992 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
4993 'only_matching': True,
4994 }, {
4995 'note': 'A channel that is not live. Should raise error',
4996 'url': 'https://www.youtube.com/user/numberphile/live',
4997 'only_matching': True,
4998 }, {
4999 'url': 'https://www.youtube.com/feed/trending',
5000 'only_matching': True,
5001 }, {
5002 'url': 'https://www.youtube.com/feed/library',
5003 'only_matching': True,
5004 }, {
5005 'url': 'https://www.youtube.com/feed/history',
5006 'only_matching': True,
5007 }, {
5008 'url': 'https://www.youtube.com/feed/subscriptions',
5009 'only_matching': True,
5010 }, {
5011 'url': 'https://www.youtube.com/feed/watch_later',
5012 'only_matching': True,
5013 }, {
5014 'note': 'Recommended - redirects to home page.',
5015 'url': 'https://www.youtube.com/feed/recommended',
5016 'only_matching': True,
5017 }, {
5018 'note': 'inline playlist with not always working continuations',
5019 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5020 'only_matching': True,
5021 }, {
5022 'url': 'https://www.youtube.com/course',
5023 'only_matching': True,
5024 }, {
5025 'url': 'https://www.youtube.com/zsecurity',
5026 'only_matching': True,
5027 }, {
5028 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5029 'only_matching': True,
5030 }, {
5031 'url': 'https://www.youtube.com/TheYoungTurks/live',
5032 'only_matching': True,
5033 }, {
5034 'url': 'https://www.youtube.com/hashtag/cctv9',
5035 'info_dict': {
5036 'id': 'cctv9',
5037 'title': '#cctv9',
5038 'tags': [],
5039 },
5040 'playlist_mincount': 350,
5041 }, {
5042 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5043 'only_matching': True,
5044 }, {
5045 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5046 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5047 'only_matching': True
5048 }, {
5049 'note': '/browse/ should redirect to /channel/',
5050 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5051 'only_matching': True
5052 }, {
5053 'note': 'VLPL, should redirect to playlist?list=PL...',
5054 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5055 'info_dict': {
5056 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5057 'uploader': 'NoCopyrightSounds',
5058 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5059 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5060 'title': 'NCS Releases',
5061 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5062 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5063 'modified_date': r're:\d{8}',
5064 'view_count': int,
5065 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5066 'tags': [],
5067 'channel': 'NoCopyrightSounds',
5068 },
5069 'playlist_mincount': 166,
5070 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5071 }, {
5072 'note': 'Topic, should redirect to playlist?list=UU...',
5073 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5074 'info_dict': {
5075 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5076 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5077 'title': 'Uploads from Royalty Free Music - Topic',
5078 'uploader': 'Royalty Free Music - Topic',
5079 'tags': [],
5080 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5081 'channel': 'Royalty Free Music - Topic',
5082 'view_count': int,
5083 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5084 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5085 'modified_date': r're:\d{8}',
5086 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5087 'description': '',
5088 },
5089 'expected_warnings': [
5090 'The URL does not have a videos tab',
5091 r'[Uu]navailable videos (are|will be) hidden',
5092 ],
5093 'playlist_mincount': 101,
5094 }, {
5095 'note': 'Topic without a UU playlist',
5096 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5097 'info_dict': {
5098 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5099 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5100 'tags': [],
5101 },
5102 'expected_warnings': [
5103 'the playlist redirect gave error',
5104 ],
5105 'playlist_mincount': 9,
5106 }, {
5107 'note': 'Youtube music Album',
5108 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5109 'info_dict': {
5110 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5111 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5112 'tags': [],
5113 'view_count': int,
5114 'description': '',
5115 'availability': 'unlisted',
5116 'modified_date': r're:\d{8}',
5117 },
5118 'playlist_count': 50,
5119 }, {
5120 'note': 'unlisted single video playlist',
5121 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5122 'info_dict': {
5123 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5124 'uploader': 'colethedj',
5125 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5126 'title': 'yt-dlp unlisted playlist test',
5127 'availability': 'unlisted',
5128 'tags': [],
5129 'modified_date': '20211208',
5130 'channel': 'colethedj',
5131 'view_count': int,
5132 'description': '',
5133 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5134 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5135 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5136 },
5137 'playlist_count': 1,
5138 }, {
5139 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5140 'url': 'https://www.youtube.com/feed/recommended',
5141 'info_dict': {
5142 'id': 'recommended',
5143 'title': 'recommended',
5144 'tags': [],
5145 },
5146 'playlist_mincount': 50,
5147 'params': {
5148 'skip_download': True,
5149 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5150 },
5151 }, {
5152 'note': 'API Fallback: /videos tab, sorted by oldest first',
5153 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5154 'info_dict': {
5155 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5156 'title': 'Cody\'sLab - Videos',
5157 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5158 'uploader': 'Cody\'sLab',
5159 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5160 'channel': 'Cody\'sLab',
5161 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5162 'tags': [],
5163 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5164 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5165 'channel_follower_count': int
5166 },
5167 'playlist_mincount': 650,
5168 'params': {
5169 'skip_download': True,
5170 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5171 },
5172 }, {
5173 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5174 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5175 'info_dict': {
5176 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5177 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5178 'title': 'Uploads from Royalty Free Music - Topic',
5179 'uploader': 'Royalty Free Music - Topic',
5180 'modified_date': r're:\d{8}',
5181 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5182 'description': '',
5183 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5184 'tags': [],
5185 'channel': 'Royalty Free Music - Topic',
5186 'view_count': int,
5187 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5188 },
5189 'expected_warnings': [
5190 'does not have a videos tab',
5191 r'[Uu]navailable videos (are|will be) hidden',
5192 ],
5193 'playlist_mincount': 101,
5194 'params': {
5195 'skip_download': True,
5196 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5197 },
5198 }, {
5199 'note': 'non-standard redirect to regional channel',
5200 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5201 'only_matching': True
5202 }, {
5203 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5204 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5205 'info_dict': {
5206 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5207 'modified_date': '20220407',
5208 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5209 'tags': [],
5210 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5211 'uploader': 'pukkandan',
5212 'availability': 'unlisted',
5213 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5214 'channel': 'pukkandan',
5215 'description': 'Test for collaborative playlist',
5216 'title': 'yt-dlp test - collaborative playlist',
5217 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5218 },
5219 'playlist_mincount': 2
5220 }]
5221
5222 @classmethod
5223 def suitable(cls, url):
5224 return False if YoutubeIE.suitable(url) else super().suitable(url)
5225
5226 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5227
5228 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5229 def _real_extract(self, url, smuggled_data):
5230 item_id = self._match_id(url)
5231 url = compat_urlparse.urlunparse(
5232 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5233 compat_opts = self.get_param('compat_opts', [])
5234
5235 def get_mobj(url):
5236 mobj = self._URL_RE.match(url).groupdict()
5237 mobj.update((k, '') for k, v in mobj.items() if v is None)
5238 return mobj
5239
5240 mobj, redirect_warning = get_mobj(url), None
5241 # Youtube returns incomplete data if tabname is not lower case
5242 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5243 if is_channel:
5244 if smuggled_data.get('is_music_url'):
5245 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5246 item_id = item_id[2:]
5247 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5248 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5249 mdata = self._extract_tab_endpoint(
5250 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5251 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5252 get_all=False, expected_type=compat_str)
5253 if not murl:
5254 raise ExtractorError('Failed to resolve album to playlist')
5255 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5256 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5257 pre = f'https://www.youtube.com/channel/{item_id}'
5258
5259 original_tab_name = tab
5260 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5261 # Home URLs should redirect to /videos/
5262 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5263 'To download only the videos in the home page, add a "/featured" to the URL')
5264 tab = '/videos'
5265
5266 url = ''.join((pre, tab, post))
5267 mobj = get_mobj(url)
5268
5269 # Handle both video/playlist URLs
5270 qs = parse_qs(url)
5271 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5272
5273 if not video_id and mobj['not_channel'].startswith('watch'):
5274 if not playlist_id:
5275 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5276 raise ExtractorError('Unable to recognize tab page')
5277 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5278 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5279 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5280 mobj = get_mobj(url)
5281
5282 if video_id and playlist_id:
5283 if self.get_param('noplaylist'):
5284 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5285 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5286 ie=YoutubeIE.ie_key(), video_id=video_id)
5287 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5288
5289 data, ytcfg = self._extract_data(url, item_id)
5290
5291 # YouTube may provide a non-standard redirect to the regional channel
5292 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5293 redirect_url = traverse_obj(
5294 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5295 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5296 redirect_url = ''.join((
5297 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5298 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5299 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5300
5301 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5302 if tabs:
5303 selected_tab = self._extract_selected_tab(tabs)
5304 selected_tab_name = selected_tab.get('title', '').lower()
5305 if selected_tab_name == 'home':
5306 selected_tab_name = 'featured'
5307 requested_tab_name = mobj['tab'][1:]
5308 if 'no-youtube-channel-redirect' not in compat_opts:
5309 if requested_tab_name == 'live':
5310 # Live tab should have redirected to the video
5311 raise ExtractorError('The channel is not currently live', expected=True)
5312 if requested_tab_name not in ('', selected_tab_name):
5313 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5314 if not original_tab_name:
5315 if item_id[:2] == 'UC':
5316 # Topic channels don't have /videos. Use the equivalent playlist instead
5317 pl_id = f'UU{item_id[2:]}'
5318 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5319 try:
5320 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5321 except ExtractorError:
5322 redirect_warning += ' and the playlist redirect gave error'
5323 else:
5324 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5325 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5326 if selected_tab_name and selected_tab_name != requested_tab_name:
5327 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5328 else:
5329 raise ExtractorError(redirect_warning, expected=True)
5330
5331 if redirect_warning:
5332 self.to_screen(redirect_warning)
5333 self.write_debug(f'Final URL: {url}')
5334
5335 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5336 if 'no-youtube-unavailable-videos' not in compat_opts:
5337 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5338 self._extract_and_report_alerts(data, only_once=True)
5339 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5340 if tabs:
5341 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5342
5343 playlist = traverse_obj(
5344 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5345 if playlist:
5346 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5347
5348 video_id = traverse_obj(
5349 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5350 if video_id:
5351 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5352 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5353 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5354 ie=YoutubeIE.ie_key(), video_id=video_id)
5355
5356 raise ExtractorError('Unable to recognize tab page')
5357
5358
5359 class YoutubePlaylistIE(InfoExtractor):
5360 IE_DESC = 'YouTube playlists'
5361 _VALID_URL = r'''(?x)(?:
5362 (?:https?://)?
5363 (?:\w+\.)?
5364 (?:
5365 (?:
5366 youtube(?:kids)?\.com|
5367 %(invidious)s
5368 )
5369 /.*?\?.*?\blist=
5370 )?
5371 (?P<id>%(playlist_id)s)
5372 )''' % {
5373 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5374 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5375 }
5376 IE_NAME = 'youtube:playlist'
5377 _TESTS = [{
5378 'note': 'issue #673',
5379 'url': 'PLBB231211A4F62143',
5380 'info_dict': {
5381 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5382 'id': 'PLBB231211A4F62143',
5383 'uploader': 'Wickman',
5384 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5385 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5386 'view_count': int,
5387 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5388 'modified_date': r're:\d{8}',
5389 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5390 'channel': 'Wickman',
5391 'tags': [],
5392 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5393 },
5394 'playlist_mincount': 29,
5395 }, {
5396 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5397 'info_dict': {
5398 'title': 'YDL_safe_search',
5399 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5400 },
5401 'playlist_count': 2,
5402 'skip': 'This playlist is private',
5403 }, {
5404 'note': 'embedded',
5405 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5406 'playlist_count': 4,
5407 'info_dict': {
5408 'title': 'JODA15',
5409 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5410 'uploader': 'milan',
5411 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5412 'description': '',
5413 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5414 'tags': [],
5415 'modified_date': '20140919',
5416 'view_count': int,
5417 'channel': 'milan',
5418 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5419 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5420 },
5421 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5422 }, {
5423 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5424 'playlist_mincount': 654,
5425 'info_dict': {
5426 'title': '2018 Chinese New Singles (11/6 updated)',
5427 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5428 'uploader': 'LBK',
5429 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5430 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5431 'channel': 'LBK',
5432 'view_count': int,
5433 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5434 'tags': [],
5435 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5436 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5437 'modified_date': r're:\d{8}',
5438 },
5439 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5440 }, {
5441 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5442 'only_matching': True,
5443 }, {
5444 # music album playlist
5445 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5446 'only_matching': True,
5447 }]
5448
5449 @classmethod
5450 def suitable(cls, url):
5451 if YoutubeTabIE.suitable(url):
5452 return False
5453 from ..utils import parse_qs
5454 qs = parse_qs(url)
5455 if qs.get('v', [None])[0]:
5456 return False
5457 return super().suitable(url)
5458
5459 def _real_extract(self, url):
5460 playlist_id = self._match_id(url)
5461 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5462 url = update_url_query(
5463 'https://www.youtube.com/playlist',
5464 parse_qs(url) or {'list': playlist_id})
5465 if is_music_url:
5466 url = smuggle_url(url, {'is_music_url': True})
5467 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5468
5469
5470 class YoutubeYtBeIE(InfoExtractor):
5471 IE_DESC = 'youtu.be'
5472 _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}
5473 _TESTS = [{
5474 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5475 'info_dict': {
5476 'id': 'yeWKywCrFtk',
5477 'ext': 'mp4',
5478 'title': 'Small Scale Baler and Braiding Rugs',
5479 'uploader': 'Backus-Page House Museum',
5480 'uploader_id': 'backuspagemuseum',
5481 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5482 'upload_date': '20161008',
5483 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5484 'categories': ['Nonprofits & Activism'],
5485 'tags': list,
5486 'like_count': int,
5487 'age_limit': 0,
5488 'playable_in_embed': True,
5489 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5490 'channel': 'Backus-Page House Museum',
5491 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5492 'live_status': 'not_live',
5493 'view_count': int,
5494 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5495 'availability': 'public',
5496 'duration': 59,
5497 },
5498 'params': {
5499 'noplaylist': True,
5500 'skip_download': True,
5501 },
5502 }, {
5503 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5504 'only_matching': True,
5505 }]
5506
5507 def _real_extract(self, url):
5508 mobj = self._match_valid_url(url)
5509 video_id = mobj.group('id')
5510 playlist_id = mobj.group('playlist_id')
5511 return self.url_result(
5512 update_url_query('https://www.youtube.com/watch', {
5513 'v': video_id,
5514 'list': playlist_id,
5515 'feature': 'youtu.be',
5516 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5517
5518
5519 class YoutubeLivestreamEmbedIE(InfoExtractor):
5520 IE_DESC = 'YouTube livestream embeds'
5521 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5522 _TESTS = [{
5523 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5524 'only_matching': True,
5525 }]
5526
5527 def _real_extract(self, url):
5528 channel_id = self._match_id(url)
5529 return self.url_result(
5530 f'https://www.youtube.com/channel/{channel_id}/live',
5531 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5532
5533
5534 class YoutubeYtUserIE(InfoExtractor):
5535 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5536 IE_NAME = 'youtube:user'
5537 _VALID_URL = r'ytuser:(?P<id>.+)'
5538 _TESTS = [{
5539 'url': 'ytuser:phihag',
5540 'only_matching': True,
5541 }]
5542
5543 def _real_extract(self, url):
5544 user_id = self._match_id(url)
5545 return self.url_result(
5546 'https://www.youtube.com/user/%s/videos' % user_id,
5547 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5548
5549
5550 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5551 IE_NAME = 'youtube:favorites'
5552 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5553 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5554 _LOGIN_REQUIRED = True
5555 _TESTS = [{
5556 'url': ':ytfav',
5557 'only_matching': True,
5558 }, {
5559 'url': ':ytfavorites',
5560 'only_matching': True,
5561 }]
5562
5563 def _real_extract(self, url):
5564 return self.url_result(
5565 'https://www.youtube.com/playlist?list=LL',
5566 ie=YoutubeTabIE.ie_key())
5567
5568
5569 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5570 IE_NAME = 'youtube:notif'
5571 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5572 _VALID_URL = r':ytnotif(?:ication)?s?'
5573 _LOGIN_REQUIRED = True
5574 _TESTS = [{
5575 'url': ':ytnotif',
5576 'only_matching': True,
5577 }, {
5578 'url': ':ytnotifications',
5579 'only_matching': True,
5580 }]
5581
5582 def _extract_notification_menu(self, response, continuation_list):
5583 notification_list = traverse_obj(
5584 response,
5585 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5586 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5587 expected_type=list) or []
5588 continuation_list[0] = None
5589 for item in notification_list:
5590 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5591 if entry:
5592 yield entry
5593 continuation = item.get('continuationItemRenderer')
5594 if continuation:
5595 continuation_list[0] = continuation
5596
5597 def _extract_notification_renderer(self, notification):
5598 video_id = traverse_obj(
5599 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5600 url = f'https://www.youtube.com/watch?v={video_id}'
5601 channel_id = None
5602 if not video_id:
5603 browse_ep = traverse_obj(
5604 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5605 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5606 post_id = self._search_regex(
5607 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5608 'post id', default=None)
5609 if not channel_id or not post_id:
5610 return
5611 # The direct /post url redirects to this in the browser
5612 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5613
5614 channel = traverse_obj(
5615 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5616 expected_type=str)
5617 title = self._search_regex(
5618 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5619 'video title', default=None)
5620 if title:
5621 title = title.replace('\xad', '') # remove soft hyphens
5622 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5623 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5624 else None)
5625 return {
5626 '_type': 'url',
5627 'url': url,
5628 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5629 'video_id': video_id,
5630 'title': title,
5631 'channel_id': channel_id,
5632 'channel': channel,
5633 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5634 'upload_date': upload_date,
5635 }
5636
5637 def _notification_menu_entries(self, ytcfg):
5638 continuation_list = [None]
5639 response = None
5640 for page in itertools.count(1):
5641 ctoken = traverse_obj(
5642 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5643 response = self._extract_response(
5644 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5645 ep='notification/get_notification_menu', check_get_keys='actions',
5646 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5647 yield from self._extract_notification_menu(response, continuation_list)
5648 if not continuation_list[0]:
5649 break
5650
5651 def _real_extract(self, url):
5652 display_id = 'notifications'
5653 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5654 self._report_playlist_authcheck(ytcfg)
5655 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5656
5657
5658 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5659 IE_DESC = 'YouTube search'
5660 IE_NAME = 'youtube:search'
5661 _SEARCH_KEY = 'ytsearch'
5662 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5663 _TESTS = [{
5664 'url': 'ytsearch5:youtube-dl test video',
5665 'playlist_count': 5,
5666 'info_dict': {
5667 'id': 'youtube-dl test video',
5668 'title': 'youtube-dl test video',
5669 }
5670 }]
5671
5672
5673 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5674 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5675 _SEARCH_KEY = 'ytsearchdate'
5676 IE_DESC = 'YouTube search, newest videos first'
5677 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5678 _TESTS = [{
5679 'url': 'ytsearchdate5:youtube-dl test video',
5680 'playlist_count': 5,
5681 'info_dict': {
5682 'id': 'youtube-dl test video',
5683 'title': 'youtube-dl test video',
5684 }
5685 }]
5686
5687
5688 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5689 IE_DESC = 'YouTube search URLs with sorting and filter support'
5690 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5691 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5692 _TESTS = [{
5693 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5694 'playlist_mincount': 5,
5695 'info_dict': {
5696 'id': 'youtube-dl test video',
5697 'title': 'youtube-dl test video',
5698 }
5699 }, {
5700 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5701 'playlist_mincount': 5,
5702 'info_dict': {
5703 'id': 'python',
5704 'title': 'python',
5705 }
5706 }, {
5707 'url': 'https://www.youtube.com/results?search_query=%23cats',
5708 'playlist_mincount': 1,
5709 'info_dict': {
5710 'id': '#cats',
5711 'title': '#cats',
5712 'entries': [{
5713 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5714 'title': '#cats',
5715 }],
5716 },
5717 }, {
5718 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5719 'only_matching': True,
5720 }]
5721
5722 def _real_extract(self, url):
5723 qs = parse_qs(url)
5724 query = (qs.get('search_query') or qs.get('q'))[0]
5725 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5726
5727
5728 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5729 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5730 IE_NAME = 'youtube:music:search_url'
5731 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5732 _TESTS = [{
5733 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5734 'playlist_count': 16,
5735 'info_dict': {
5736 'id': 'royalty free music',
5737 'title': 'royalty free music',
5738 }
5739 }, {
5740 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5741 'playlist_mincount': 30,
5742 'info_dict': {
5743 'id': 'royalty free music - songs',
5744 'title': 'royalty free music - songs',
5745 },
5746 'params': {'extract_flat': 'in_playlist'}
5747 }, {
5748 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5749 'playlist_mincount': 30,
5750 'info_dict': {
5751 'id': 'royalty free music - community playlists',
5752 'title': 'royalty free music - community playlists',
5753 },
5754 'params': {'extract_flat': 'in_playlist'}
5755 }]
5756
5757 _SECTIONS = {
5758 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5759 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5760 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5761 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5762 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5763 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5764 }
5765
5766 def _real_extract(self, url):
5767 qs = parse_qs(url)
5768 query = (qs.get('search_query') or qs.get('q'))[0]
5769 params = qs.get('sp', (None,))[0]
5770 if params:
5771 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5772 else:
5773 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5774 params = self._SECTIONS.get(section)
5775 if not params:
5776 section = None
5777 title = join_nonempty(query, section, delim=' - ')
5778 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5779
5780
5781 class YoutubeFeedsInfoExtractor(InfoExtractor):
5782 """
5783 Base class for feed extractors
5784 Subclasses must define the _FEED_NAME property.
5785 """
5786 _LOGIN_REQUIRED = True
5787
5788 def _real_initialize(self):
5789 YoutubeBaseInfoExtractor._check_login_required(self)
5790
5791 @property
5792 def IE_NAME(self):
5793 return 'youtube:%s' % self._FEED_NAME
5794
5795 def _real_extract(self, url):
5796 return self.url_result(
5797 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5798
5799
5800 class YoutubeWatchLaterIE(InfoExtractor):
5801 IE_NAME = 'youtube:watchlater'
5802 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5803 _VALID_URL = r':ytwatchlater'
5804 _TESTS = [{
5805 'url': ':ytwatchlater',
5806 'only_matching': True,
5807 }]
5808
5809 def _real_extract(self, url):
5810 return self.url_result(
5811 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5812
5813
5814 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5815 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5816 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5817 _FEED_NAME = 'recommended'
5818 _LOGIN_REQUIRED = False
5819 _TESTS = [{
5820 'url': ':ytrec',
5821 'only_matching': True,
5822 }, {
5823 'url': ':ytrecommended',
5824 'only_matching': True,
5825 }, {
5826 'url': 'https://youtube.com',
5827 'only_matching': True,
5828 }]
5829
5830
5831 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5832 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5833 _VALID_URL = r':ytsub(?:scription)?s?'
5834 _FEED_NAME = 'subscriptions'
5835 _TESTS = [{
5836 'url': ':ytsubs',
5837 'only_matching': True,
5838 }, {
5839 'url': ':ytsubscriptions',
5840 'only_matching': True,
5841 }]
5842
5843
5844 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5845 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5846 _VALID_URL = r':ythis(?:tory)?'
5847 _FEED_NAME = 'history'
5848 _TESTS = [{
5849 'url': ':ythistory',
5850 'only_matching': True,
5851 }]
5852
5853
5854 class YoutubeStoriesIE(InfoExtractor):
5855 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5856 IE_NAME = 'youtube:stories'
5857 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5858 _TESTS = [{
5859 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5860 'only_matching': True,
5861 }]
5862
5863 def _real_extract(self, url):
5864 playlist_id = f'RLTD{self._match_id(url)}'
5865 return self.url_result(
5866 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5867 ie=YoutubeTabIE, video_id=playlist_id)
5868
5869
5870 class YoutubeTruncatedURLIE(InfoExtractor):
5871 IE_NAME = 'youtube:truncated_url'
5872 IE_DESC = False # Do not list
5873 _VALID_URL = r'''(?x)
5874 (?:https?://)?
5875 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5876 (?:watch\?(?:
5877 feature=[a-z_]+|
5878 annotation_id=annotation_[^&]+|
5879 x-yt-cl=[0-9]+|
5880 hl=[^&]*|
5881 t=[0-9]+
5882 )?
5883 |
5884 attribution_link\?a=[^&]+
5885 )
5886 $
5887 '''
5888
5889 _TESTS = [{
5890 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5891 'only_matching': True,
5892 }, {
5893 'url': 'https://www.youtube.com/watch?',
5894 'only_matching': True,
5895 }, {
5896 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5897 'only_matching': True,
5898 }, {
5899 'url': 'https://www.youtube.com/watch?feature=foo',
5900 'only_matching': True,
5901 }, {
5902 'url': 'https://www.youtube.com/watch?hl=en-GB',
5903 'only_matching': True,
5904 }, {
5905 'url': 'https://www.youtube.com/watch?t=2372',
5906 'only_matching': True,
5907 }]
5908
5909 def _real_extract(self, url):
5910 raise ExtractorError(
5911 'Did you forget to quote the URL? Remember that & is a meta '
5912 'character in most shells, so you want to put the URL in quotes, '
5913 'like youtube-dl '
5914 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5915 ' or simply youtube-dl BaW_jenozKc .',
5916 expected=True)
5917
5918
5919 class YoutubeClipIE(InfoExtractor):
5920 IE_NAME = 'youtube:clip'
5921 IE_DESC = False # Do not list
5922 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5923
5924 def _real_extract(self, url):
5925 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5926 return self.url_result(url, 'Generic')
5927
5928
5929 class YoutubeTruncatedIDIE(InfoExtractor):
5930 IE_NAME = 'youtube:truncated_id'
5931 IE_DESC = False # Do not list
5932 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5933
5934 _TESTS = [{
5935 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5936 'only_matching': True,
5937 }]
5938
5939 def _real_extract(self, url):
5940 video_id = self._match_id(url)
5941 raise ExtractorError(
5942 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5943 expected=True)