]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[toggo] Improve `_VALID_URL` (#3689)
[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(f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3177 dct = {
3178 'asr': int_or_none(fmt.get('audioSampleRate')),
3179 'filesize': int_or_none(fmt.get('contentLength')),
3180 'format_id': itag,
3181 'format_note': join_nonempty(
3182 '%s%s' % (audio_track.get('displayName') or '',
3183 ' (default)' if language_preference > 0 else ''),
3184 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3185 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3186 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3187 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3188 'fps': int_or_none(fmt.get('fps')) or None,
3189 'height': height,
3190 'quality': q(quality),
3191 'has_drm': bool(fmt.get('drmFamilies')),
3192 'tbr': tbr,
3193 'url': fmt_url,
3194 'width': int_or_none(fmt.get('width')),
3195 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3196 'desc' if language_preference < -1 else ''),
3197 'language_preference': language_preference,
3198 # Strictly de-prioritize damaged and 3gp formats
3199 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3200 }
3201 mime_mobj = re.match(
3202 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3203 if mime_mobj:
3204 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3205 dct.update(parse_codecs(mime_mobj.group(2)))
3206 no_audio = dct.get('acodec') == 'none'
3207 no_video = dct.get('vcodec') == 'none'
3208 if no_audio:
3209 dct['vbr'] = tbr
3210 if no_video:
3211 dct['abr'] = tbr
3212 if no_audio or no_video:
3213 dct['downloader_options'] = {
3214 # Youtube throttles chunks >~10M
3215 'http_chunk_size': 10485760,
3216 }
3217 if dct.get('ext'):
3218 dct['container'] = dct['ext'] + '_dash'
3219 yield dct
3220
3221 live_from_start = is_live and self.get_param('live_from_start')
3222 skip_manifests = self._configuration_arg('skip')
3223 if not self.get_param('youtube_include_hls_manifest', True):
3224 skip_manifests.append('hls')
3225 get_dash = 'dash' not in skip_manifests and (
3226 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3227 get_hls = not live_from_start and 'hls' not in skip_manifests
3228
3229 def process_manifest_format(f, proto, itag):
3230 if itag in itags:
3231 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3232 return False
3233 itag = f'{itag}-{proto}'
3234 if itag:
3235 f['format_id'] = itag
3236 itags[itag] = proto
3237
3238 f['quality'] = next((
3239 q(qdict[val])
3240 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3241 if val in qdict), -1)
3242 return True
3243
3244 for sd in streaming_data:
3245 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3246 if hls_manifest_url:
3247 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3248 if process_manifest_format(f, 'hls', self._search_regex(
3249 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3250 yield f
3251
3252 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3253 if dash_manifest_url:
3254 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3255 if process_manifest_format(f, 'dash', f['format_id']):
3256 f['filesize'] = int_or_none(self._search_regex(
3257 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3258 if live_from_start:
3259 f['is_from_start'] = True
3260
3261 yield f
3262
3263 def _extract_storyboard(self, player_responses, duration):
3264 spec = get_first(
3265 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3266 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3267 if not base_url:
3268 return
3269 L = len(spec) - 1
3270 for i, args in enumerate(spec):
3271 args = args.split('#')
3272 counts = list(map(int_or_none, args[:5]))
3273 if len(args) != 8 or not all(counts):
3274 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3275 continue
3276 width, height, frame_count, cols, rows = counts
3277 N, sigh = args[6:]
3278
3279 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3280 fragment_count = frame_count / (cols * rows)
3281 fragment_duration = duration / fragment_count
3282 yield {
3283 'format_id': f'sb{i}',
3284 'format_note': 'storyboard',
3285 'ext': 'mhtml',
3286 'protocol': 'mhtml',
3287 'acodec': 'none',
3288 'vcodec': 'none',
3289 'url': url,
3290 'width': width,
3291 'height': height,
3292 'fragments': [{
3293 'url': url.replace('$M', str(j)),
3294 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3295 } for j in range(math.ceil(fragment_count))],
3296 }
3297
3298 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3299 webpage = None
3300 if 'webpage' not in self._configuration_arg('player_skip'):
3301 webpage = self._download_webpage(
3302 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3303
3304 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3305
3306 player_responses, player_url = self._extract_player_responses(
3307 self._get_requested_clients(url, smuggled_data),
3308 video_id, webpage, master_ytcfg)
3309
3310 return webpage, master_ytcfg, player_responses, player_url
3311
3312 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3313 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3314 is_live = get_first(video_details, 'isLive')
3315 if is_live is None:
3316 is_live = get_first(live_broadcast_details, 'isLiveNow')
3317
3318 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3319 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3320
3321 return live_broadcast_details, is_live, streaming_data, formats
3322
3323 def _real_extract(self, url):
3324 url, smuggled_data = unsmuggle_url(url, {})
3325 video_id = self._match_id(url)
3326
3327 base_url = self.http_scheme() + '//www.youtube.com/'
3328 webpage_url = base_url + 'watch?v=' + video_id
3329
3330 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3331
3332 playability_statuses = traverse_obj(
3333 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3334
3335 trailer_video_id = get_first(
3336 playability_statuses,
3337 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3338 expected_type=str)
3339 if trailer_video_id:
3340 return self.url_result(
3341 trailer_video_id, self.ie_key(), trailer_video_id)
3342
3343 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3344 if webpage else (lambda x: None))
3345
3346 video_details = traverse_obj(
3347 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3348 microformats = traverse_obj(
3349 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3350 expected_type=dict, default=[])
3351 video_title = (
3352 get_first(video_details, 'title')
3353 or self._get_text(microformats, (..., 'title'))
3354 or search_meta(['og:title', 'twitter:title', 'title']))
3355 video_description = get_first(video_details, 'shortDescription')
3356
3357 multifeed_metadata_list = get_first(
3358 player_responses,
3359 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3360 expected_type=str)
3361 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3362 if self.get_param('noplaylist'):
3363 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3364 else:
3365 entries = []
3366 feed_ids = []
3367 for feed in multifeed_metadata_list.split(','):
3368 # Unquote should take place before split on comma (,) since textual
3369 # fields may contain comma as well (see
3370 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3371 feed_data = compat_parse_qs(
3372 compat_urllib_parse_unquote_plus(feed))
3373
3374 def feed_entry(name):
3375 return try_get(
3376 feed_data, lambda x: x[name][0], compat_str)
3377
3378 feed_id = feed_entry('id')
3379 if not feed_id:
3380 continue
3381 feed_title = feed_entry('title')
3382 title = video_title
3383 if feed_title:
3384 title += ' (%s)' % feed_title
3385 entries.append({
3386 '_type': 'url_transparent',
3387 'ie_key': 'Youtube',
3388 'url': smuggle_url(
3389 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3390 {'force_singlefeed': True}),
3391 'title': title,
3392 })
3393 feed_ids.append(feed_id)
3394 self.to_screen(
3395 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3396 % (', '.join(feed_ids), video_id))
3397 return self.playlist_result(
3398 entries, video_id, video_title, video_description)
3399
3400 duration = int_or_none(
3401 get_first(video_details, 'lengthSeconds')
3402 or get_first(microformats, 'lengthSeconds')
3403 or parse_duration(search_meta('duration'))) or None
3404
3405 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3406 video_id, microformats, video_details, player_responses, player_url, duration)
3407
3408 if not formats:
3409 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3410 self.report_drm(video_id)
3411 pemr = get_first(
3412 playability_statuses,
3413 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3414 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3415 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3416 if subreason:
3417 if subreason == 'The uploader has not made this video available in your country.':
3418 countries = get_first(microformats, 'availableCountries')
3419 if not countries:
3420 regions_allowed = search_meta('regionsAllowed')
3421 countries = regions_allowed.split(',') if regions_allowed else None
3422 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3423 reason += f'. {subreason}'
3424 if reason:
3425 self.raise_no_formats(reason, expected=True)
3426
3427 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3428 if not keywords and webpage:
3429 keywords = [
3430 unescapeHTML(m.group('content'))
3431 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3432 for keyword in keywords:
3433 if keyword.startswith('yt:stretch='):
3434 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3435 if mobj:
3436 # NB: float is intentional for forcing float division
3437 w, h = (float(v) for v in mobj.groups())
3438 if w > 0 and h > 0:
3439 ratio = w / h
3440 for f in formats:
3441 if f.get('vcodec') != 'none':
3442 f['stretched_ratio'] = ratio
3443 break
3444 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3445 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3446 if thumbnail_url:
3447 thumbnails.append({
3448 'url': thumbnail_url,
3449 })
3450 original_thumbnails = thumbnails.copy()
3451
3452 # The best resolution thumbnails sometimes does not appear in the webpage
3453 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3454 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3455 thumbnail_names = [
3456 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3457 # in resolution, these are not the custom thumbnail. So de-prioritize them
3458 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3459 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3460 ]
3461 n_thumbnail_names = len(thumbnail_names)
3462 thumbnails.extend({
3463 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3464 video_id=video_id, name=name, ext=ext,
3465 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3466 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3467 for thumb in thumbnails:
3468 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3469 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3470 self._remove_duplicate_formats(thumbnails)
3471 self._downloader._sort_thumbnails(original_thumbnails)
3472
3473 category = get_first(microformats, 'category') or search_meta('genre')
3474 channel_id = str_or_none(
3475 get_first(video_details, 'channelId')
3476 or get_first(microformats, 'externalChannelId')
3477 or search_meta('channelId'))
3478 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3479
3480 live_content = get_first(video_details, 'isLiveContent')
3481 is_upcoming = get_first(video_details, 'isUpcoming')
3482 if is_live is None:
3483 if is_upcoming or live_content is False:
3484 is_live = False
3485 if is_upcoming is None and (live_content or is_live):
3486 is_upcoming = False
3487 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3488 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3489 if not duration and live_end_time and live_start_time:
3490 duration = live_end_time - live_start_time
3491
3492 if is_live and self.get_param('live_from_start'):
3493 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3494
3495 formats.extend(self._extract_storyboard(player_responses, duration))
3496
3497 # Source is given priority since formats that throttle are given lower source_preference
3498 # When throttling issue is fully fixed, remove this
3499 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3500
3501 info = {
3502 'id': video_id,
3503 'title': video_title,
3504 'formats': formats,
3505 'thumbnails': thumbnails,
3506 # The best thumbnail that we are sure exists. Prevents unnecessary
3507 # URL checking if user don't care about getting the best possible thumbnail
3508 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3509 'description': video_description,
3510 'uploader': get_first(video_details, 'author'),
3511 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3512 'uploader_url': owner_profile_url,
3513 'channel_id': channel_id,
3514 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3515 'duration': duration,
3516 'view_count': int_or_none(
3517 get_first((video_details, microformats), (..., 'viewCount'))
3518 or search_meta('interactionCount')),
3519 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3520 'age_limit': 18 if (
3521 get_first(microformats, 'isFamilySafe') is False
3522 or search_meta('isFamilyFriendly') == 'false'
3523 or search_meta('og:restrictions:age') == '18+') else 0,
3524 'webpage_url': webpage_url,
3525 'categories': [category] if category else None,
3526 'tags': keywords,
3527 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3528 'is_live': is_live,
3529 'was_live': (False if is_live or is_upcoming or live_content is False
3530 else None if is_live is None or is_upcoming is None
3531 else live_content),
3532 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3533 'release_timestamp': live_start_time,
3534 }
3535
3536 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3537 if pctr:
3538 def get_lang_code(track):
3539 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3540 or track.get('languageCode'))
3541
3542 # Converted into dicts to remove duplicates
3543 captions = {
3544 get_lang_code(sub): sub
3545 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3546 translation_languages = {
3547 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3548 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3549
3550 def process_language(container, base_url, lang_code, sub_name, query):
3551 lang_subs = container.setdefault(lang_code, [])
3552 for fmt in self._SUBTITLE_FORMATS:
3553 query.update({
3554 'fmt': fmt,
3555 })
3556 lang_subs.append({
3557 'ext': fmt,
3558 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3559 'name': sub_name,
3560 })
3561
3562 subtitles, automatic_captions = {}, {}
3563 for lang_code, caption_track in captions.items():
3564 base_url = caption_track.get('baseUrl')
3565 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3566 if not base_url:
3567 continue
3568 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3569 if caption_track.get('kind') != 'asr':
3570 if not lang_code:
3571 continue
3572 process_language(
3573 subtitles, base_url, lang_code, lang_name, {})
3574 if not caption_track.get('isTranslatable'):
3575 continue
3576 for trans_code, trans_name in translation_languages.items():
3577 if not trans_code:
3578 continue
3579 orig_trans_code = trans_code
3580 if caption_track.get('kind') != 'asr':
3581 if 'translated_subs' in self._configuration_arg('skip'):
3582 continue
3583 trans_code += f'-{lang_code}'
3584 trans_name += format_field(lang_name, template=' from %s')
3585 # Add an "-orig" label to the original language so that it can be distinguished.
3586 # The subs are returned without "-orig" as well for compatibility
3587 if lang_code == f'a-{orig_trans_code}':
3588 process_language(
3589 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3590 # Setting tlang=lang returns damaged subtitles.
3591 process_language(automatic_captions, base_url, trans_code, trans_name,
3592 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3593 info['automatic_captions'] = automatic_captions
3594 info['subtitles'] = subtitles
3595
3596 parsed_url = compat_urllib_parse_urlparse(url)
3597 for component in [parsed_url.fragment, parsed_url.query]:
3598 query = compat_parse_qs(component)
3599 for k, v in query.items():
3600 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3601 d_k += '_time'
3602 if d_k not in info and k in s_ks:
3603 info[d_k] = parse_duration(query[k][0])
3604
3605 # Youtube Music Auto-generated description
3606 if video_description:
3607 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)
3608 if mobj:
3609 release_year = mobj.group('release_year')
3610 release_date = mobj.group('release_date')
3611 if release_date:
3612 release_date = release_date.replace('-', '')
3613 if not release_year:
3614 release_year = release_date[:4]
3615 info.update({
3616 'album': mobj.group('album'.strip()),
3617 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3618 'track': mobj.group('track').strip(),
3619 'release_date': release_date,
3620 'release_year': int_or_none(release_year),
3621 })
3622
3623 initial_data = None
3624 if webpage:
3625 initial_data = self._extract_yt_initial_variable(
3626 webpage, self._YT_INITIAL_DATA_RE, video_id,
3627 'yt initial data')
3628 if not initial_data:
3629 query = {'videoId': video_id}
3630 query.update(self._get_checkok_params())
3631 initial_data = self._extract_response(
3632 item_id=video_id, ep='next', fatal=False,
3633 ytcfg=master_ytcfg, query=query,
3634 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3635 note='Downloading initial data API JSON')
3636
3637 try: # This will error if there is no livechat
3638 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3639 except (KeyError, IndexError, TypeError):
3640 pass
3641 else:
3642 info.setdefault('subtitles', {})['live_chat'] = [{
3643 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3644 'video_id': video_id,
3645 'ext': 'json',
3646 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3647 }]
3648
3649 if initial_data:
3650 info['chapters'] = (
3651 self._extract_chapters_from_json(initial_data, duration)
3652 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3653 or None)
3654
3655 contents = traverse_obj(
3656 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3657 expected_type=list, default=[])
3658
3659 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3660 if vpir:
3661 stl = vpir.get('superTitleLink')
3662 if stl:
3663 stl = self._get_text(stl)
3664 if try_get(
3665 vpir,
3666 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3667 info['location'] = stl
3668 else:
3669 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3670 if mobj:
3671 info.update({
3672 'series': mobj.group(1),
3673 'season_number': int(mobj.group(2)),
3674 'episode_number': int(mobj.group(3)),
3675 })
3676 for tlb in (try_get(
3677 vpir,
3678 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3679 list) or []):
3680 tbr = tlb.get('toggleButtonRenderer') or {}
3681 for getter, regex in [(
3682 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3683 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3684 lambda x: x['accessibility'],
3685 lambda x: x['accessibilityData']['accessibilityData'],
3686 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3687 label = (try_get(tbr, getter, dict) or {}).get('label')
3688 if label:
3689 mobj = re.match(regex, label)
3690 if mobj:
3691 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3692 break
3693 sbr_tooltip = try_get(
3694 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3695 if sbr_tooltip:
3696 like_count, dislike_count = sbr_tooltip.split(' / ')
3697 info.update({
3698 'like_count': str_to_int(like_count),
3699 'dislike_count': str_to_int(dislike_count),
3700 })
3701 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3702 if vsir:
3703 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3704 info.update({
3705 'channel': self._get_text(vor, 'title'),
3706 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3707
3708 rows = try_get(
3709 vsir,
3710 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3711 list) or []
3712 multiple_songs = False
3713 for row in rows:
3714 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3715 multiple_songs = True
3716 break
3717 for row in rows:
3718 mrr = row.get('metadataRowRenderer') or {}
3719 mrr_title = mrr.get('title')
3720 if not mrr_title:
3721 continue
3722 mrr_title = self._get_text(mrr, 'title')
3723 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3724 if mrr_title == 'License':
3725 info['license'] = mrr_contents_text
3726 elif not multiple_songs:
3727 if mrr_title == 'Album':
3728 info['album'] = mrr_contents_text
3729 elif mrr_title == 'Artist':
3730 info['artist'] = mrr_contents_text
3731 elif mrr_title == 'Song':
3732 info['track'] = mrr_contents_text
3733
3734 fallbacks = {
3735 'channel': 'uploader',
3736 'channel_id': 'uploader_id',
3737 'channel_url': 'uploader_url',
3738 }
3739
3740 # The upload date for scheduled, live and past live streams / premieres in microformats
3741 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3742 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3743 upload_date = (
3744 unified_strdate(get_first(microformats, 'uploadDate'))
3745 or unified_strdate(search_meta('uploadDate')))
3746 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3747 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3748 info['upload_date'] = upload_date
3749
3750 for to, frm in fallbacks.items():
3751 if not info.get(to):
3752 info[to] = info.get(frm)
3753
3754 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3755 v = info.get(s_k)
3756 if v:
3757 info[d_k] = v
3758
3759 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3760 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3761 is_membersonly = None
3762 is_premium = None
3763 if initial_data and is_private is not None:
3764 is_membersonly = False
3765 is_premium = False
3766 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3767 badge_labels = set()
3768 for content in contents:
3769 if not isinstance(content, dict):
3770 continue
3771 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3772 for badge_label in badge_labels:
3773 if badge_label.lower() == 'members only':
3774 is_membersonly = True
3775 elif badge_label.lower() == 'premium':
3776 is_premium = True
3777 elif badge_label.lower() == 'unlisted':
3778 is_unlisted = True
3779
3780 info['availability'] = self._availability(
3781 is_private=is_private,
3782 needs_premium=is_premium,
3783 needs_subscription=is_membersonly,
3784 needs_auth=info['age_limit'] >= 18,
3785 is_unlisted=None if is_private is None else is_unlisted)
3786
3787 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3788
3789 self.mark_watched(video_id, player_responses)
3790
3791 return info
3792
3793
3794 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3795
3796 @staticmethod
3797 def passthrough_smuggled_data(func):
3798 def _smuggle(entries, smuggled_data):
3799 for entry in entries:
3800 # TODO: Convert URL to music.youtube instead.
3801 # Do we need to passthrough any other smuggled_data?
3802 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3803 yield entry
3804
3805 @functools.wraps(func)
3806 def wrapper(self, url):
3807 url, smuggled_data = unsmuggle_url(url, {})
3808 if self.is_music_url(url):
3809 smuggled_data['is_music_url'] = True
3810 info_dict = func(self, url, smuggled_data)
3811 if smuggled_data and info_dict.get('entries'):
3812 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3813 return info_dict
3814 return wrapper
3815
3816 def _extract_channel_id(self, webpage):
3817 channel_id = self._html_search_meta(
3818 'channelId', webpage, 'channel id', default=None)
3819 if channel_id:
3820 return channel_id
3821 channel_url = self._html_search_meta(
3822 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3823 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3824 'twitter:app:url:googleplay'), webpage, 'channel url')
3825 return self._search_regex(
3826 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3827 channel_url, 'channel id')
3828
3829 @staticmethod
3830 def _extract_basic_item_renderer(item):
3831 # Modified from _extract_grid_item_renderer
3832 known_basic_renderers = (
3833 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3834 )
3835 for key, renderer in item.items():
3836 if not isinstance(renderer, dict):
3837 continue
3838 elif key in known_basic_renderers:
3839 return renderer
3840 elif key.startswith('grid') and key.endswith('Renderer'):
3841 return renderer
3842
3843 def _grid_entries(self, grid_renderer):
3844 for item in grid_renderer['items']:
3845 if not isinstance(item, dict):
3846 continue
3847 renderer = self._extract_basic_item_renderer(item)
3848 if not isinstance(renderer, dict):
3849 continue
3850 title = self._get_text(renderer, 'title')
3851
3852 # playlist
3853 playlist_id = renderer.get('playlistId')
3854 if playlist_id:
3855 yield self.url_result(
3856 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3857 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3858 video_title=title)
3859 continue
3860 # video
3861 video_id = renderer.get('videoId')
3862 if video_id:
3863 yield self._extract_video(renderer)
3864 continue
3865 # channel
3866 channel_id = renderer.get('channelId')
3867 if channel_id:
3868 yield self.url_result(
3869 'https://www.youtube.com/channel/%s' % channel_id,
3870 ie=YoutubeTabIE.ie_key(), video_title=title)
3871 continue
3872 # generic endpoint URL support
3873 ep_url = urljoin('https://www.youtube.com/', try_get(
3874 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3875 compat_str))
3876 if ep_url:
3877 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3878 if ie.suitable(ep_url):
3879 yield self.url_result(
3880 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3881 break
3882
3883 def _music_reponsive_list_entry(self, renderer):
3884 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3885 if video_id:
3886 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3887 ie=YoutubeIE.ie_key(), video_id=video_id)
3888 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3889 if playlist_id:
3890 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3891 if video_id:
3892 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3893 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3894 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3895 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3896 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3897 if browse_id:
3898 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3899 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3900
3901 def _shelf_entries_from_content(self, shelf_renderer):
3902 content = shelf_renderer.get('content')
3903 if not isinstance(content, dict):
3904 return
3905 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3906 if renderer:
3907 # TODO: add support for nested playlists so each shelf is processed
3908 # as separate playlist
3909 # TODO: this includes only first N items
3910 yield from self._grid_entries(renderer)
3911 renderer = content.get('horizontalListRenderer')
3912 if renderer:
3913 # TODO
3914 pass
3915
3916 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3917 ep = try_get(
3918 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3919 compat_str)
3920 shelf_url = urljoin('https://www.youtube.com', ep)
3921 if shelf_url:
3922 # Skipping links to another channels, note that checking for
3923 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3924 # will not work
3925 if skip_channels and '/channels?' in shelf_url:
3926 return
3927 title = self._get_text(shelf_renderer, 'title')
3928 yield self.url_result(shelf_url, video_title=title)
3929 # Shelf may not contain shelf URL, fallback to extraction from content
3930 yield from self._shelf_entries_from_content(shelf_renderer)
3931
3932 def _playlist_entries(self, video_list_renderer):
3933 for content in video_list_renderer['contents']:
3934 if not isinstance(content, dict):
3935 continue
3936 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3937 if not isinstance(renderer, dict):
3938 continue
3939 video_id = renderer.get('videoId')
3940 if not video_id:
3941 continue
3942 yield self._extract_video(renderer)
3943
3944 def _rich_entries(self, rich_grid_renderer):
3945 renderer = try_get(
3946 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3947 video_id = renderer.get('videoId')
3948 if not video_id:
3949 return
3950 yield self._extract_video(renderer)
3951
3952 def _video_entry(self, video_renderer):
3953 video_id = video_renderer.get('videoId')
3954 if video_id:
3955 return self._extract_video(video_renderer)
3956
3957 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3958 url = urljoin('https://youtube.com', traverse_obj(
3959 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3960 if url:
3961 return self.url_result(
3962 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3963
3964 def _post_thread_entries(self, post_thread_renderer):
3965 post_renderer = try_get(
3966 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3967 if not post_renderer:
3968 return
3969 # video attachment
3970 video_renderer = try_get(
3971 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3972 video_id = video_renderer.get('videoId')
3973 if video_id:
3974 entry = self._extract_video(video_renderer)
3975 if entry:
3976 yield entry
3977 # playlist attachment
3978 playlist_id = try_get(
3979 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3980 if playlist_id:
3981 yield self.url_result(
3982 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3983 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3984 # inline video links
3985 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
3986 for run in runs:
3987 if not isinstance(run, dict):
3988 continue
3989 ep_url = try_get(
3990 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
3991 if not ep_url:
3992 continue
3993 if not YoutubeIE.suitable(ep_url):
3994 continue
3995 ep_video_id = YoutubeIE._match_id(ep_url)
3996 if video_id == ep_video_id:
3997 continue
3998 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
3999
4000 def _post_thread_continuation_entries(self, post_thread_continuation):
4001 contents = post_thread_continuation.get('contents')
4002 if not isinstance(contents, list):
4003 return
4004 for content in contents:
4005 renderer = content.get('backstagePostThreadRenderer')
4006 if not isinstance(renderer, dict):
4007 continue
4008 yield from self._post_thread_entries(renderer)
4009
4010 r''' # unused
4011 def _rich_grid_entries(self, contents):
4012 for content in contents:
4013 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4014 if video_renderer:
4015 entry = self._video_entry(video_renderer)
4016 if entry:
4017 yield entry
4018 '''
4019
4020 def _extract_entries(self, parent_renderer, continuation_list):
4021 # continuation_list is modified in-place with continuation_list = [continuation_token]
4022 continuation_list[:] = [None]
4023 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4024 for content in contents:
4025 if not isinstance(content, dict):
4026 continue
4027 is_renderer = traverse_obj(
4028 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4029 expected_type=dict)
4030 if not is_renderer:
4031 renderer = content.get('richItemRenderer')
4032 if renderer:
4033 for entry in self._rich_entries(renderer):
4034 yield entry
4035 continuation_list[0] = self._extract_continuation(parent_renderer)
4036 continue
4037 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4038 for isr_content in isr_contents:
4039 if not isinstance(isr_content, dict):
4040 continue
4041
4042 known_renderers = {
4043 'playlistVideoListRenderer': self._playlist_entries,
4044 'gridRenderer': self._grid_entries,
4045 'reelShelfRenderer': self._grid_entries,
4046 'shelfRenderer': self._shelf_entries,
4047 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4048 'backstagePostThreadRenderer': self._post_thread_entries,
4049 'videoRenderer': lambda x: [self._video_entry(x)],
4050 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4051 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4052 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4053 }
4054 for key, renderer in isr_content.items():
4055 if key not in known_renderers:
4056 continue
4057 for entry in known_renderers[key](renderer):
4058 if entry:
4059 yield entry
4060 continuation_list[0] = self._extract_continuation(renderer)
4061 break
4062
4063 if not continuation_list[0]:
4064 continuation_list[0] = self._extract_continuation(is_renderer)
4065
4066 if not continuation_list[0]:
4067 continuation_list[0] = self._extract_continuation(parent_renderer)
4068
4069 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4070 continuation_list = [None]
4071 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4072 tab_content = try_get(tab, lambda x: x['content'], dict)
4073 if not tab_content:
4074 return
4075 parent_renderer = (
4076 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4077 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4078 yield from extract_entries(parent_renderer)
4079 continuation = continuation_list[0]
4080
4081 for page_num in itertools.count(1):
4082 if not continuation:
4083 break
4084 headers = self.generate_api_headers(
4085 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4086 response = self._extract_response(
4087 item_id=f'{item_id} page {page_num}',
4088 query=continuation, headers=headers, ytcfg=ytcfg,
4089 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4090
4091 if not response:
4092 break
4093 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4094 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4095 visitor_data = self._extract_visitor_data(response) or visitor_data
4096
4097 known_continuation_renderers = {
4098 'playlistVideoListContinuation': self._playlist_entries,
4099 'gridContinuation': self._grid_entries,
4100 'itemSectionContinuation': self._post_thread_continuation_entries,
4101 'sectionListContinuation': extract_entries, # for feeds
4102 }
4103 continuation_contents = try_get(
4104 response, lambda x: x['continuationContents'], dict) or {}
4105 continuation_renderer = None
4106 for key, value in continuation_contents.items():
4107 if key not in known_continuation_renderers:
4108 continue
4109 continuation_renderer = value
4110 continuation_list = [None]
4111 yield from known_continuation_renderers[key](continuation_renderer)
4112 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4113 break
4114 if continuation_renderer:
4115 continue
4116
4117 known_renderers = {
4118 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4119 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4120 'gridVideoRenderer': (self._grid_entries, 'items'),
4121 'gridChannelRenderer': (self._grid_entries, 'items'),
4122 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4123 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4124 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4125 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4126 }
4127 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4128 continuation_items = try_get(
4129 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4130 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4131 video_items_renderer = None
4132 for key, value in continuation_item.items():
4133 if key not in known_renderers:
4134 continue
4135 video_items_renderer = {known_renderers[key][1]: continuation_items}
4136 continuation_list = [None]
4137 yield from known_renderers[key][0](video_items_renderer)
4138 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4139 break
4140 if video_items_renderer:
4141 continue
4142 break
4143
4144 @staticmethod
4145 def _extract_selected_tab(tabs, fatal=True):
4146 for tab in tabs:
4147 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4148 if renderer.get('selected') is True:
4149 return renderer
4150 else:
4151 if fatal:
4152 raise ExtractorError('Unable to find selected tab')
4153
4154 def _extract_uploader(self, data):
4155 uploader = {}
4156 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4157 owner = try_get(
4158 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4159 if owner:
4160 owner_text = owner.get('text')
4161 uploader['uploader'] = self._search_regex(
4162 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4163 uploader['uploader_id'] = try_get(
4164 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4165 uploader['uploader_url'] = urljoin(
4166 'https://www.youtube.com/',
4167 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4168 return {k: v for k, v in uploader.items() if v is not None}
4169
4170 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4171 playlist_id = title = description = channel_url = channel_name = channel_id = None
4172 tags = []
4173
4174 selected_tab = self._extract_selected_tab(tabs)
4175 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4176 renderer = try_get(
4177 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4178 if renderer:
4179 channel_name = renderer.get('title')
4180 channel_url = renderer.get('channelUrl')
4181 channel_id = renderer.get('externalId')
4182 else:
4183 renderer = try_get(
4184 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4185
4186 if renderer:
4187 title = renderer.get('title')
4188 description = renderer.get('description', '')
4189 playlist_id = channel_id
4190 tags = renderer.get('keywords', '').split()
4191
4192 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4193 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4194 def _get_uncropped(url):
4195 return url_or_none((url or '').split('=')[0] + '=s0')
4196
4197 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4198 if avatar_thumbnails:
4199 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4200 if uncropped_avatar:
4201 avatar_thumbnails.append({
4202 'url': uncropped_avatar,
4203 'id': 'avatar_uncropped',
4204 'preference': 1
4205 })
4206
4207 channel_banners = self._extract_thumbnails(
4208 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4209 for banner in channel_banners:
4210 banner['preference'] = -10
4211
4212 if channel_banners:
4213 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4214 if uncropped_banner:
4215 channel_banners.append({
4216 'url': uncropped_banner,
4217 'id': 'banner_uncropped',
4218 'preference': -5
4219 })
4220
4221 primary_thumbnails = self._extract_thumbnails(
4222 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4223
4224 if playlist_id is None:
4225 playlist_id = item_id
4226
4227 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4228 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4229 if title is None:
4230 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4231 title += format_field(selected_tab, 'title', ' - %s')
4232 title += format_field(selected_tab, 'expandedText', ' - %s')
4233
4234 metadata = {
4235 'playlist_id': playlist_id,
4236 'playlist_title': title,
4237 'playlist_description': description,
4238 'uploader': channel_name,
4239 'uploader_id': channel_id,
4240 'uploader_url': channel_url,
4241 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4242 'tags': tags,
4243 'view_count': self._get_count(playlist_stats, 1),
4244 'availability': self._extract_availability(data),
4245 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4246 'playlist_count': self._get_count(playlist_stats, 0),
4247 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4248 }
4249 if not channel_id:
4250 metadata.update(self._extract_uploader(data))
4251 metadata.update({
4252 'channel': metadata['uploader'],
4253 'channel_id': metadata['uploader_id'],
4254 'channel_url': metadata['uploader_url']})
4255 return self.playlist_result(
4256 self._entries(
4257 selected_tab, playlist_id, ytcfg,
4258 self._extract_account_syncid(ytcfg, data),
4259 self._extract_visitor_data(data, ytcfg)),
4260 **metadata)
4261
4262 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4263 first_id = last_id = response = None
4264 for page_num in itertools.count(1):
4265 videos = list(self._playlist_entries(playlist))
4266 if not videos:
4267 return
4268 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4269 if start >= len(videos):
4270 return
4271 for video in videos[start:]:
4272 yield video
4273 first_id = first_id or videos[0]['id']
4274 last_id = videos[-1]['id']
4275 watch_endpoint = try_get(
4276 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4277 headers = self.generate_api_headers(
4278 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4279 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4280 query = {
4281 'playlistId': playlist_id,
4282 'videoId': watch_endpoint.get('videoId') or last_id,
4283 'index': watch_endpoint.get('index') or len(videos),
4284 'params': watch_endpoint.get('params') or 'OAE%3D'
4285 }
4286 response = self._extract_response(
4287 item_id='%s page %d' % (playlist_id, page_num),
4288 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4289 check_get_keys='contents'
4290 )
4291 playlist = try_get(
4292 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4293
4294 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4295 title = playlist.get('title') or try_get(
4296 data, lambda x: x['titleText']['simpleText'], compat_str)
4297 playlist_id = playlist.get('playlistId') or item_id
4298
4299 # Delegating everything except mix playlists to regular tab-based playlist URL
4300 playlist_url = urljoin(url, try_get(
4301 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4302 compat_str))
4303
4304 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4305 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4306 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4307
4308 if playlist_url and playlist_url != url and not is_known_unviewable:
4309 return self.url_result(
4310 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4311 video_title=title)
4312
4313 return self.playlist_result(
4314 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4315 playlist_id=playlist_id, playlist_title=title)
4316
4317 def _extract_availability(self, data):
4318 """
4319 Gets the availability of a given playlist/tab.
4320 Note: Unless YouTube tells us explicitly, we do not assume it is public
4321 @param data: response
4322 """
4323 is_private = is_unlisted = None
4324 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4325 badge_labels = self._extract_badges(renderer)
4326
4327 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4328 privacy_dropdown_entries = try_get(
4329 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4330 for renderer_dict in privacy_dropdown_entries:
4331 is_selected = try_get(
4332 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4333 if not is_selected:
4334 continue
4335 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4336 if label:
4337 badge_labels.add(label.lower())
4338 break
4339
4340 for badge_label in badge_labels:
4341 if badge_label == 'unlisted':
4342 is_unlisted = True
4343 elif badge_label == 'private':
4344 is_private = True
4345 elif badge_label == 'public':
4346 is_unlisted = is_private = False
4347 return self._availability(is_private, False, False, False, is_unlisted)
4348
4349 @staticmethod
4350 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4351 sidebar_renderer = try_get(
4352 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4353 for item in sidebar_renderer:
4354 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4355 if renderer:
4356 return renderer
4357
4358 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4359 """
4360 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4361 """
4362 browse_id = params = None
4363 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4364 if not renderer:
4365 return
4366 menu_renderer = try_get(
4367 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4368 for menu_item in menu_renderer:
4369 if not isinstance(menu_item, dict):
4370 continue
4371 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4372 text = try_get(
4373 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4374 if not text or text.lower() != 'show unavailable videos':
4375 continue
4376 browse_endpoint = try_get(
4377 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4378 browse_id = browse_endpoint.get('browseId')
4379 params = browse_endpoint.get('params')
4380 break
4381
4382 headers = self.generate_api_headers(
4383 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4384 visitor_data=self._extract_visitor_data(data, ytcfg))
4385 query = {
4386 'params': params or 'wgYCCAA=',
4387 'browseId': browse_id or 'VL%s' % item_id
4388 }
4389 return self._extract_response(
4390 item_id=item_id, headers=headers, query=query,
4391 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4392 note='Downloading API JSON with unavailable videos')
4393
4394 @property
4395 def skip_webpage(self):
4396 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4397
4398 def _extract_webpage(self, url, item_id, fatal=True):
4399 retries = self.get_param('extractor_retries', 3)
4400 count = -1
4401 webpage = data = last_error = None
4402 while count < retries:
4403 count += 1
4404 # Sometimes youtube returns a webpage with incomplete ytInitialData
4405 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4406 if last_error:
4407 self.report_warning('%s. Retrying ...' % last_error)
4408 try:
4409 webpage = self._download_webpage(
4410 url, item_id,
4411 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4412 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4413 except ExtractorError as e:
4414 if isinstance(e.cause, network_exceptions):
4415 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4416 last_error = error_to_compat_str(e.cause or e.msg)
4417 if count < retries:
4418 continue
4419 if fatal:
4420 raise
4421 self.report_warning(error_to_compat_str(e))
4422 break
4423 else:
4424 try:
4425 self._extract_and_report_alerts(data)
4426 except ExtractorError as e:
4427 if fatal:
4428 raise
4429 self.report_warning(error_to_compat_str(e))
4430 break
4431
4432 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4433 break
4434
4435 last_error = 'Incomplete yt initial data received'
4436 if count >= retries:
4437 if fatal:
4438 raise ExtractorError(last_error)
4439 self.report_warning(last_error)
4440 break
4441
4442 return webpage, data
4443
4444 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4445 """Use if failed to extract ytcfg (and data) from initial webpage"""
4446 if not ytcfg and self.is_authenticated:
4447 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4448 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4449 raise ExtractorError(
4450 f'{msg}. If you are not downloading private content, or '
4451 'your cookies are only for the first account and channel,'
4452 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4453 expected=True)
4454 self.report_warning(msg, only_once=True)
4455
4456 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4457 data = None
4458 if not self.skip_webpage:
4459 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4460 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4461 # Reject webpage data if redirected to home page without explicitly requesting
4462 selected_tab = self._extract_selected_tab(traverse_obj(
4463 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4464 if (url != 'https://www.youtube.com/feed/recommended'
4465 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4466 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4467 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4468 if fatal:
4469 raise ExtractorError(msg, expected=True)
4470 self.report_warning(msg, only_once=True)
4471 if not data:
4472 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4473 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4474 return data, ytcfg
4475
4476 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4477 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4478 resolve_response = self._extract_response(
4479 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4480 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4481 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4482 for ep_key, ep in endpoints.items():
4483 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4484 if params:
4485 return self._extract_response(
4486 item_id=item_id, query=params, ep=ep, headers=headers,
4487 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4488 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4489 err_note = 'Failed to resolve url (does the playlist exist?)'
4490 if fatal:
4491 raise ExtractorError(err_note, expected=True)
4492 self.report_warning(err_note, item_id)
4493
4494 _SEARCH_PARAMS = None
4495
4496 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4497 data = {'query': query}
4498 if params is NO_DEFAULT:
4499 params = self._SEARCH_PARAMS
4500 if params:
4501 data['params'] = params
4502
4503 content_keys = (
4504 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4505 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4506 # ytmusic search
4507 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4508 ('continuationContents', ),
4509 )
4510 display_id = f'query "{query}"'
4511 check_get_keys = tuple({keys[0] for keys in content_keys})
4512 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4513 self._report_playlist_authcheck(ytcfg, fatal=False)
4514
4515 continuation_list = [None]
4516 search = None
4517 for page_num in itertools.count(1):
4518 data.update(continuation_list[0] or {})
4519 headers = self.generate_api_headers(
4520 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4521 search = self._extract_response(
4522 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4523 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4524 slr_contents = traverse_obj(search, *content_keys)
4525 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4526 if not continuation_list[0]:
4527 break
4528
4529
4530 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4531 IE_DESC = 'YouTube Tabs'
4532 _VALID_URL = r'''(?x:
4533 https?://
4534 (?:\w+\.)?
4535 (?:
4536 youtube(?:kids)?\.com|
4537 %(invidious)s
4538 )/
4539 (?:
4540 (?P<channel_type>channel|c|user|browse)/|
4541 (?P<not_channel>
4542 feed/|hashtag/|
4543 (?:playlist|watch)\?.*?\blist=
4544 )|
4545 (?!(?:%(reserved_names)s)\b) # Direct URLs
4546 )
4547 (?P<id>[^/?\#&]+)
4548 )''' % {
4549 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4550 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4551 }
4552 IE_NAME = 'youtube:tab'
4553
4554 _TESTS = [{
4555 'note': 'playlists, multipage',
4556 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4557 'playlist_mincount': 94,
4558 'info_dict': {
4559 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4560 'title': 'Igor Kleiner - Playlists',
4561 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4562 'uploader': 'Igor Kleiner',
4563 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4564 'channel': 'Igor Kleiner',
4565 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4566 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4567 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4568 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4569 'channel_follower_count': int
4570 },
4571 }, {
4572 'note': 'playlists, multipage, different order',
4573 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4574 'playlist_mincount': 94,
4575 'info_dict': {
4576 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4577 'title': 'Igor Kleiner - Playlists',
4578 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4579 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4580 'uploader': 'Igor Kleiner',
4581 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4582 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4583 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4584 'channel': 'Igor Kleiner',
4585 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4586 'channel_follower_count': int
4587 },
4588 }, {
4589 'note': 'playlists, series',
4590 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4591 'playlist_mincount': 5,
4592 'info_dict': {
4593 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4594 'title': '3Blue1Brown - Playlists',
4595 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4596 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4597 'uploader': '3Blue1Brown',
4598 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4599 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4600 'channel': '3Blue1Brown',
4601 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4602 'tags': ['Mathematics'],
4603 'channel_follower_count': int
4604 },
4605 }, {
4606 'note': 'playlists, singlepage',
4607 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4608 'playlist_mincount': 4,
4609 'info_dict': {
4610 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4611 'title': 'ThirstForScience - Playlists',
4612 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4613 'uploader': 'ThirstForScience',
4614 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4615 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4616 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4617 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4618 'tags': 'count:13',
4619 'channel': 'ThirstForScience',
4620 'channel_follower_count': int
4621 }
4622 }, {
4623 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4624 'only_matching': True,
4625 }, {
4626 'note': 'basic, single video playlist',
4627 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4628 'info_dict': {
4629 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4630 'uploader': 'Sergey M.',
4631 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4632 'title': 'youtube-dl public playlist',
4633 'description': '',
4634 'tags': [],
4635 'view_count': int,
4636 'modified_date': '20201130',
4637 'channel': 'Sergey M.',
4638 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4639 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4640 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4641 },
4642 'playlist_count': 1,
4643 }, {
4644 'note': 'empty playlist',
4645 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4646 'info_dict': {
4647 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4648 'uploader': 'Sergey M.',
4649 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4650 'title': 'youtube-dl empty playlist',
4651 'tags': [],
4652 'channel': 'Sergey M.',
4653 'description': '',
4654 'modified_date': '20160902',
4655 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4656 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4657 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4658 },
4659 'playlist_count': 0,
4660 }, {
4661 'note': 'Home tab',
4662 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4663 'info_dict': {
4664 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4665 'title': 'lex will - Home',
4666 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4667 'uploader': 'lex will',
4668 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4669 'channel': 'lex will',
4670 'tags': ['bible', 'history', 'prophesy'],
4671 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4672 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4673 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4674 'channel_follower_count': int
4675 },
4676 'playlist_mincount': 2,
4677 }, {
4678 'note': 'Videos tab',
4679 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4680 'info_dict': {
4681 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4682 'title': 'lex will - Videos',
4683 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4684 'uploader': 'lex will',
4685 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4686 'tags': ['bible', 'history', 'prophesy'],
4687 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4688 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4689 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4690 'channel': 'lex will',
4691 'channel_follower_count': int
4692 },
4693 'playlist_mincount': 975,
4694 }, {
4695 'note': 'Videos tab, sorted by popular',
4696 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4697 'info_dict': {
4698 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4699 'title': 'lex will - Videos',
4700 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4701 'uploader': 'lex will',
4702 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4703 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4704 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4705 'channel': 'lex will',
4706 'tags': ['bible', 'history', 'prophesy'],
4707 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4708 'channel_follower_count': int
4709 },
4710 'playlist_mincount': 199,
4711 }, {
4712 'note': 'Playlists tab',
4713 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4714 'info_dict': {
4715 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4716 'title': 'lex will - Playlists',
4717 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4718 'uploader': 'lex will',
4719 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4720 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4721 'channel': 'lex will',
4722 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4723 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4724 'tags': ['bible', 'history', 'prophesy'],
4725 'channel_follower_count': int
4726 },
4727 'playlist_mincount': 17,
4728 }, {
4729 'note': 'Community tab',
4730 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4731 'info_dict': {
4732 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4733 'title': 'lex will - Community',
4734 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4735 'uploader': 'lex will',
4736 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4737 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4738 'channel': 'lex will',
4739 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4740 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'tags': ['bible', 'history', 'prophesy'],
4742 'channel_follower_count': int
4743 },
4744 'playlist_mincount': 18,
4745 }, {
4746 'note': 'Channels tab',
4747 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4748 'info_dict': {
4749 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4750 'title': 'lex will - Channels',
4751 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4752 'uploader': 'lex will',
4753 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4754 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4755 'channel': 'lex will',
4756 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4757 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4758 'tags': ['bible', 'history', 'prophesy'],
4759 'channel_follower_count': int
4760 },
4761 'playlist_mincount': 12,
4762 }, {
4763 'note': 'Search tab',
4764 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4765 'playlist_mincount': 40,
4766 'info_dict': {
4767 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4768 'title': '3Blue1Brown - Search - linear algebra',
4769 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4770 'uploader': '3Blue1Brown',
4771 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4772 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4773 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4774 'tags': ['Mathematics'],
4775 'channel': '3Blue1Brown',
4776 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4777 'channel_follower_count': int
4778 },
4779 }, {
4780 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4781 'only_matching': True,
4782 }, {
4783 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4784 'only_matching': True,
4785 }, {
4786 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4787 'only_matching': True,
4788 }, {
4789 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4790 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4791 'info_dict': {
4792 'title': '29C3: Not my department',
4793 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4794 'uploader': 'Christiaan008',
4795 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4796 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4797 'tags': [],
4798 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4799 'view_count': int,
4800 'modified_date': '20150605',
4801 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4802 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4803 'channel': 'Christiaan008',
4804 },
4805 'playlist_count': 96,
4806 }, {
4807 'note': 'Large playlist',
4808 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4809 'info_dict': {
4810 'title': 'Uploads from Cauchemar',
4811 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4812 'uploader': 'Cauchemar',
4813 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4814 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4815 'tags': [],
4816 'modified_date': r're:\d{8}',
4817 'channel': 'Cauchemar',
4818 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4819 'view_count': int,
4820 'description': '',
4821 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4822 },
4823 'playlist_mincount': 1123,
4824 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4825 }, {
4826 'note': 'even larger playlist, 8832 videos',
4827 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4828 'only_matching': True,
4829 }, {
4830 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4831 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4832 'info_dict': {
4833 'title': 'Uploads from Interstellar Movie',
4834 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4835 'uploader': 'Interstellar Movie',
4836 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4837 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4838 'tags': [],
4839 'view_count': int,
4840 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4841 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4842 'channel': 'Interstellar Movie',
4843 'description': '',
4844 'modified_date': r're:\d{8}',
4845 },
4846 'playlist_mincount': 21,
4847 }, {
4848 'note': 'Playlist with "show unavailable videos" button',
4849 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4850 'info_dict': {
4851 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4852 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4853 'uploader': 'Phim Siêu Nhân Nhật Bản',
4854 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4855 'view_count': int,
4856 'channel': 'Phim Siêu Nhân Nhật Bản',
4857 'tags': [],
4858 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4859 'description': '',
4860 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4861 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4862 'modified_date': r're:\d{8}',
4863 },
4864 'playlist_mincount': 200,
4865 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4866 }, {
4867 'note': 'Playlist with unavailable videos in page 7',
4868 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4869 'info_dict': {
4870 'title': 'Uploads from BlankTV',
4871 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4872 'uploader': 'BlankTV',
4873 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4874 'channel': 'BlankTV',
4875 'channel_url': 'https://www.youtube.com/c/blanktv',
4876 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4877 'view_count': int,
4878 'tags': [],
4879 'uploader_url': 'https://www.youtube.com/c/blanktv',
4880 'modified_date': r're:\d{8}',
4881 'description': '',
4882 },
4883 'playlist_mincount': 1000,
4884 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4885 }, {
4886 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4887 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4888 'info_dict': {
4889 'title': 'Data Analysis with Dr Mike Pound',
4890 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4891 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4892 'uploader': 'Computerphile',
4893 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4894 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4895 'tags': [],
4896 'view_count': int,
4897 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4898 'channel_url': 'https://www.youtube.com/user/Computerphile',
4899 'channel': 'Computerphile',
4900 },
4901 'playlist_mincount': 11,
4902 }, {
4903 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4904 'only_matching': True,
4905 }, {
4906 'note': 'Playlist URL that does not actually serve a playlist',
4907 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4908 'info_dict': {
4909 'id': 'FqZTN594JQw',
4910 'ext': 'webm',
4911 'title': "Smiley's People 01 detective, Adventure Series, Action",
4912 'uploader': 'STREEM',
4913 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4914 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4915 'upload_date': '20150526',
4916 'license': 'Standard YouTube License',
4917 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4918 'categories': ['People & Blogs'],
4919 'tags': list,
4920 'view_count': int,
4921 'like_count': int,
4922 },
4923 'params': {
4924 'skip_download': True,
4925 },
4926 'skip': 'This video is not available.',
4927 'add_ie': [YoutubeIE.ie_key()],
4928 }, {
4929 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4930 'only_matching': True,
4931 }, {
4932 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4933 'only_matching': True,
4934 }, {
4935 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4936 'info_dict': {
4937 'id': 'GgL890LIznQ', # This will keep changing
4938 'ext': 'mp4',
4939 'title': str,
4940 'uploader': 'Sky News',
4941 'uploader_id': 'skynews',
4942 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4943 'upload_date': r're:\d{8}',
4944 'description': str,
4945 'categories': ['News & Politics'],
4946 'tags': list,
4947 'like_count': int,
4948 'release_timestamp': 1642502819,
4949 'channel': 'Sky News',
4950 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4951 'age_limit': 0,
4952 'view_count': int,
4953 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4954 'playable_in_embed': True,
4955 'release_date': '20220118',
4956 'availability': 'public',
4957 'live_status': 'is_live',
4958 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4959 'channel_follower_count': int
4960 },
4961 'params': {
4962 'skip_download': True,
4963 },
4964 'expected_warnings': ['Ignoring subtitle tracks found in '],
4965 }, {
4966 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4967 'info_dict': {
4968 'id': 'a48o2S1cPoo',
4969 'ext': 'mp4',
4970 'title': 'The Young Turks - Live Main Show',
4971 'uploader': 'The Young Turks',
4972 'uploader_id': 'TheYoungTurks',
4973 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4974 'upload_date': '20150715',
4975 'license': 'Standard YouTube License',
4976 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4977 'categories': ['News & Politics'],
4978 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4979 'like_count': int,
4980 },
4981 'params': {
4982 'skip_download': True,
4983 },
4984 'only_matching': True,
4985 }, {
4986 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
4987 'only_matching': True,
4988 }, {
4989 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
4990 'only_matching': True,
4991 }, {
4992 'note': 'A channel that is not live. Should raise error',
4993 'url': 'https://www.youtube.com/user/numberphile/live',
4994 'only_matching': True,
4995 }, {
4996 'url': 'https://www.youtube.com/feed/trending',
4997 'only_matching': True,
4998 }, {
4999 'url': 'https://www.youtube.com/feed/library',
5000 'only_matching': True,
5001 }, {
5002 'url': 'https://www.youtube.com/feed/history',
5003 'only_matching': True,
5004 }, {
5005 'url': 'https://www.youtube.com/feed/subscriptions',
5006 'only_matching': True,
5007 }, {
5008 'url': 'https://www.youtube.com/feed/watch_later',
5009 'only_matching': True,
5010 }, {
5011 'note': 'Recommended - redirects to home page.',
5012 'url': 'https://www.youtube.com/feed/recommended',
5013 'only_matching': True,
5014 }, {
5015 'note': 'inline playlist with not always working continuations',
5016 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5017 'only_matching': True,
5018 }, {
5019 'url': 'https://www.youtube.com/course',
5020 'only_matching': True,
5021 }, {
5022 'url': 'https://www.youtube.com/zsecurity',
5023 'only_matching': True,
5024 }, {
5025 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5026 'only_matching': True,
5027 }, {
5028 'url': 'https://www.youtube.com/TheYoungTurks/live',
5029 'only_matching': True,
5030 }, {
5031 'url': 'https://www.youtube.com/hashtag/cctv9',
5032 'info_dict': {
5033 'id': 'cctv9',
5034 'title': '#cctv9',
5035 'tags': [],
5036 },
5037 'playlist_mincount': 350,
5038 }, {
5039 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5040 'only_matching': True,
5041 }, {
5042 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5043 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5044 'only_matching': True
5045 }, {
5046 'note': '/browse/ should redirect to /channel/',
5047 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5048 'only_matching': True
5049 }, {
5050 'note': 'VLPL, should redirect to playlist?list=PL...',
5051 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5052 'info_dict': {
5053 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5054 'uploader': 'NoCopyrightSounds',
5055 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5056 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5057 'title': 'NCS Releases',
5058 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5059 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5060 'modified_date': r're:\d{8}',
5061 'view_count': int,
5062 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5063 'tags': [],
5064 'channel': 'NoCopyrightSounds',
5065 },
5066 'playlist_mincount': 166,
5067 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5068 }, {
5069 'note': 'Topic, should redirect to playlist?list=UU...',
5070 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5071 'info_dict': {
5072 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5073 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5074 'title': 'Uploads from Royalty Free Music - Topic',
5075 'uploader': 'Royalty Free Music - Topic',
5076 'tags': [],
5077 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5078 'channel': 'Royalty Free Music - Topic',
5079 'view_count': int,
5080 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5081 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5082 'modified_date': r're:\d{8}',
5083 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5084 'description': '',
5085 },
5086 'expected_warnings': [
5087 'The URL does not have a videos tab',
5088 r'[Uu]navailable videos (are|will be) hidden',
5089 ],
5090 'playlist_mincount': 101,
5091 }, {
5092 'note': 'Topic without a UU playlist',
5093 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5094 'info_dict': {
5095 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5096 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5097 'tags': [],
5098 },
5099 'expected_warnings': [
5100 'the playlist redirect gave error',
5101 ],
5102 'playlist_mincount': 9,
5103 }, {
5104 'note': 'Youtube music Album',
5105 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5106 'info_dict': {
5107 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5108 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5109 'tags': [],
5110 'view_count': int,
5111 'description': '',
5112 'availability': 'unlisted',
5113 'modified_date': r're:\d{8}',
5114 },
5115 'playlist_count': 50,
5116 }, {
5117 'note': 'unlisted single video playlist',
5118 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5119 'info_dict': {
5120 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5121 'uploader': 'colethedj',
5122 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5123 'title': 'yt-dlp unlisted playlist test',
5124 'availability': 'unlisted',
5125 'tags': [],
5126 'modified_date': '20211208',
5127 'channel': 'colethedj',
5128 'view_count': int,
5129 'description': '',
5130 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5131 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5132 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5133 },
5134 'playlist_count': 1,
5135 }, {
5136 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5137 'url': 'https://www.youtube.com/feed/recommended',
5138 'info_dict': {
5139 'id': 'recommended',
5140 'title': 'recommended',
5141 'tags': [],
5142 },
5143 'playlist_mincount': 50,
5144 'params': {
5145 'skip_download': True,
5146 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5147 },
5148 }, {
5149 'note': 'API Fallback: /videos tab, sorted by oldest first',
5150 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5151 'info_dict': {
5152 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5153 'title': 'Cody\'sLab - Videos',
5154 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5155 'uploader': 'Cody\'sLab',
5156 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5157 'channel': 'Cody\'sLab',
5158 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5159 'tags': [],
5160 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5161 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5162 'channel_follower_count': int
5163 },
5164 'playlist_mincount': 650,
5165 'params': {
5166 'skip_download': True,
5167 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5168 },
5169 }, {
5170 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5171 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5172 'info_dict': {
5173 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5174 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5175 'title': 'Uploads from Royalty Free Music - Topic',
5176 'uploader': 'Royalty Free Music - Topic',
5177 'modified_date': r're:\d{8}',
5178 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5179 'description': '',
5180 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5181 'tags': [],
5182 'channel': 'Royalty Free Music - Topic',
5183 'view_count': int,
5184 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5185 },
5186 'expected_warnings': [
5187 'does not have a videos tab',
5188 r'[Uu]navailable videos (are|will be) hidden',
5189 ],
5190 'playlist_mincount': 101,
5191 'params': {
5192 'skip_download': True,
5193 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5194 },
5195 }, {
5196 'note': 'non-standard redirect to regional channel',
5197 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5198 'only_matching': True
5199 }, {
5200 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5201 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5202 'info_dict': {
5203 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5204 'modified_date': '20220407',
5205 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5206 'tags': [],
5207 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5208 'uploader': 'pukkandan',
5209 'availability': 'unlisted',
5210 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5211 'channel': 'pukkandan',
5212 'description': 'Test for collaborative playlist',
5213 'title': 'yt-dlp test - collaborative playlist',
5214 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5215 },
5216 'playlist_mincount': 2
5217 }]
5218
5219 @classmethod
5220 def suitable(cls, url):
5221 return False if YoutubeIE.suitable(url) else super().suitable(url)
5222
5223 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5224
5225 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5226 def _real_extract(self, url, smuggled_data):
5227 item_id = self._match_id(url)
5228 url = compat_urlparse.urlunparse(
5229 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5230 compat_opts = self.get_param('compat_opts', [])
5231
5232 def get_mobj(url):
5233 mobj = self._URL_RE.match(url).groupdict()
5234 mobj.update((k, '') for k, v in mobj.items() if v is None)
5235 return mobj
5236
5237 mobj, redirect_warning = get_mobj(url), None
5238 # Youtube returns incomplete data if tabname is not lower case
5239 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5240 if is_channel:
5241 if smuggled_data.get('is_music_url'):
5242 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5243 item_id = item_id[2:]
5244 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5245 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5246 mdata = self._extract_tab_endpoint(
5247 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5248 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5249 get_all=False, expected_type=compat_str)
5250 if not murl:
5251 raise ExtractorError('Failed to resolve album to playlist')
5252 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5253 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5254 pre = f'https://www.youtube.com/channel/{item_id}'
5255
5256 original_tab_name = tab
5257 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5258 # Home URLs should redirect to /videos/
5259 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5260 'To download only the videos in the home page, add a "/featured" to the URL')
5261 tab = '/videos'
5262
5263 url = ''.join((pre, tab, post))
5264 mobj = get_mobj(url)
5265
5266 # Handle both video/playlist URLs
5267 qs = parse_qs(url)
5268 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5269
5270 if not video_id and mobj['not_channel'].startswith('watch'):
5271 if not playlist_id:
5272 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5273 raise ExtractorError('Unable to recognize tab page')
5274 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5275 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5276 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5277 mobj = get_mobj(url)
5278
5279 if video_id and playlist_id:
5280 if self.get_param('noplaylist'):
5281 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5282 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5283 ie=YoutubeIE.ie_key(), video_id=video_id)
5284 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5285
5286 data, ytcfg = self._extract_data(url, item_id)
5287
5288 # YouTube may provide a non-standard redirect to the regional channel
5289 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5290 redirect_url = traverse_obj(
5291 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5292 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5293 redirect_url = ''.join((
5294 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5295 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5296 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5297
5298 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5299 if tabs:
5300 selected_tab = self._extract_selected_tab(tabs)
5301 selected_tab_name = selected_tab.get('title', '').lower()
5302 if selected_tab_name == 'home':
5303 selected_tab_name = 'featured'
5304 requested_tab_name = mobj['tab'][1:]
5305 if 'no-youtube-channel-redirect' not in compat_opts:
5306 if requested_tab_name == 'live':
5307 # Live tab should have redirected to the video
5308 raise ExtractorError('The channel is not currently live', expected=True)
5309 if requested_tab_name not in ('', selected_tab_name):
5310 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5311 if not original_tab_name:
5312 if item_id[:2] == 'UC':
5313 # Topic channels don't have /videos. Use the equivalent playlist instead
5314 pl_id = f'UU{item_id[2:]}'
5315 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5316 try:
5317 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5318 except ExtractorError:
5319 redirect_warning += ' and the playlist redirect gave error'
5320 else:
5321 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5322 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5323 if selected_tab_name and selected_tab_name != requested_tab_name:
5324 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5325 else:
5326 raise ExtractorError(redirect_warning, expected=True)
5327
5328 if redirect_warning:
5329 self.to_screen(redirect_warning)
5330 self.write_debug(f'Final URL: {url}')
5331
5332 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5333 if 'no-youtube-unavailable-videos' not in compat_opts:
5334 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5335 self._extract_and_report_alerts(data, only_once=True)
5336 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5337 if tabs:
5338 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5339
5340 playlist = traverse_obj(
5341 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5342 if playlist:
5343 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5344
5345 video_id = traverse_obj(
5346 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5347 if video_id:
5348 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5349 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5350 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5351 ie=YoutubeIE.ie_key(), video_id=video_id)
5352
5353 raise ExtractorError('Unable to recognize tab page')
5354
5355
5356 class YoutubePlaylistIE(InfoExtractor):
5357 IE_DESC = 'YouTube playlists'
5358 _VALID_URL = r'''(?x)(?:
5359 (?:https?://)?
5360 (?:\w+\.)?
5361 (?:
5362 (?:
5363 youtube(?:kids)?\.com|
5364 %(invidious)s
5365 )
5366 /.*?\?.*?\blist=
5367 )?
5368 (?P<id>%(playlist_id)s)
5369 )''' % {
5370 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5371 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5372 }
5373 IE_NAME = 'youtube:playlist'
5374 _TESTS = [{
5375 'note': 'issue #673',
5376 'url': 'PLBB231211A4F62143',
5377 'info_dict': {
5378 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5379 'id': 'PLBB231211A4F62143',
5380 'uploader': 'Wickman',
5381 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5382 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5383 'view_count': int,
5384 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5385 'modified_date': r're:\d{8}',
5386 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5387 'channel': 'Wickman',
5388 'tags': [],
5389 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5390 },
5391 'playlist_mincount': 29,
5392 }, {
5393 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5394 'info_dict': {
5395 'title': 'YDL_safe_search',
5396 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5397 },
5398 'playlist_count': 2,
5399 'skip': 'This playlist is private',
5400 }, {
5401 'note': 'embedded',
5402 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5403 'playlist_count': 4,
5404 'info_dict': {
5405 'title': 'JODA15',
5406 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5407 'uploader': 'milan',
5408 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5409 'description': '',
5410 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5411 'tags': [],
5412 'modified_date': '20140919',
5413 'view_count': int,
5414 'channel': 'milan',
5415 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5416 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5417 },
5418 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5419 }, {
5420 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5421 'playlist_mincount': 654,
5422 'info_dict': {
5423 'title': '2018 Chinese New Singles (11/6 updated)',
5424 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5425 'uploader': 'LBK',
5426 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5427 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5428 'channel': 'LBK',
5429 'view_count': int,
5430 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5431 'tags': [],
5432 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5433 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5434 'modified_date': r're:\d{8}',
5435 },
5436 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5437 }, {
5438 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5439 'only_matching': True,
5440 }, {
5441 # music album playlist
5442 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5443 'only_matching': True,
5444 }]
5445
5446 @classmethod
5447 def suitable(cls, url):
5448 if YoutubeTabIE.suitable(url):
5449 return False
5450 from ..utils import parse_qs
5451 qs = parse_qs(url)
5452 if qs.get('v', [None])[0]:
5453 return False
5454 return super().suitable(url)
5455
5456 def _real_extract(self, url):
5457 playlist_id = self._match_id(url)
5458 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5459 url = update_url_query(
5460 'https://www.youtube.com/playlist',
5461 parse_qs(url) or {'list': playlist_id})
5462 if is_music_url:
5463 url = smuggle_url(url, {'is_music_url': True})
5464 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5465
5466
5467 class YoutubeYtBeIE(InfoExtractor):
5468 IE_DESC = 'youtu.be'
5469 _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}
5470 _TESTS = [{
5471 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5472 'info_dict': {
5473 'id': 'yeWKywCrFtk',
5474 'ext': 'mp4',
5475 'title': 'Small Scale Baler and Braiding Rugs',
5476 'uploader': 'Backus-Page House Museum',
5477 'uploader_id': 'backuspagemuseum',
5478 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5479 'upload_date': '20161008',
5480 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5481 'categories': ['Nonprofits & Activism'],
5482 'tags': list,
5483 'like_count': int,
5484 'age_limit': 0,
5485 'playable_in_embed': True,
5486 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5487 'channel': 'Backus-Page House Museum',
5488 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5489 'live_status': 'not_live',
5490 'view_count': int,
5491 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5492 'availability': 'public',
5493 'duration': 59,
5494 },
5495 'params': {
5496 'noplaylist': True,
5497 'skip_download': True,
5498 },
5499 }, {
5500 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5501 'only_matching': True,
5502 }]
5503
5504 def _real_extract(self, url):
5505 mobj = self._match_valid_url(url)
5506 video_id = mobj.group('id')
5507 playlist_id = mobj.group('playlist_id')
5508 return self.url_result(
5509 update_url_query('https://www.youtube.com/watch', {
5510 'v': video_id,
5511 'list': playlist_id,
5512 'feature': 'youtu.be',
5513 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5514
5515
5516 class YoutubeLivestreamEmbedIE(InfoExtractor):
5517 IE_DESC = 'YouTube livestream embeds'
5518 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5519 _TESTS = [{
5520 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5521 'only_matching': True,
5522 }]
5523
5524 def _real_extract(self, url):
5525 channel_id = self._match_id(url)
5526 return self.url_result(
5527 f'https://www.youtube.com/channel/{channel_id}/live',
5528 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5529
5530
5531 class YoutubeYtUserIE(InfoExtractor):
5532 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5533 IE_NAME = 'youtube:user'
5534 _VALID_URL = r'ytuser:(?P<id>.+)'
5535 _TESTS = [{
5536 'url': 'ytuser:phihag',
5537 'only_matching': True,
5538 }]
5539
5540 def _real_extract(self, url):
5541 user_id = self._match_id(url)
5542 return self.url_result(
5543 'https://www.youtube.com/user/%s/videos' % user_id,
5544 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5545
5546
5547 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5548 IE_NAME = 'youtube:favorites'
5549 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5550 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5551 _LOGIN_REQUIRED = True
5552 _TESTS = [{
5553 'url': ':ytfav',
5554 'only_matching': True,
5555 }, {
5556 'url': ':ytfavorites',
5557 'only_matching': True,
5558 }]
5559
5560 def _real_extract(self, url):
5561 return self.url_result(
5562 'https://www.youtube.com/playlist?list=LL',
5563 ie=YoutubeTabIE.ie_key())
5564
5565
5566 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5567 IE_NAME = 'youtube:notif'
5568 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5569 _VALID_URL = r':ytnotif(?:ication)?s?'
5570 _LOGIN_REQUIRED = True
5571 _TESTS = [{
5572 'url': ':ytnotif',
5573 'only_matching': True,
5574 }, {
5575 'url': ':ytnotifications',
5576 'only_matching': True,
5577 }]
5578
5579 def _extract_notification_menu(self, response, continuation_list):
5580 notification_list = traverse_obj(
5581 response,
5582 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5583 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5584 expected_type=list) or []
5585 continuation_list[0] = None
5586 for item in notification_list:
5587 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5588 if entry:
5589 yield entry
5590 continuation = item.get('continuationItemRenderer')
5591 if continuation:
5592 continuation_list[0] = continuation
5593
5594 def _extract_notification_renderer(self, notification):
5595 video_id = traverse_obj(
5596 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5597 url = f'https://www.youtube.com/watch?v={video_id}'
5598 channel_id = None
5599 if not video_id:
5600 browse_ep = traverse_obj(
5601 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5602 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5603 post_id = self._search_regex(
5604 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5605 'post id', default=None)
5606 if not channel_id or not post_id:
5607 return
5608 # The direct /post url redirects to this in the browser
5609 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5610
5611 channel = traverse_obj(
5612 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5613 expected_type=str)
5614 title = self._search_regex(
5615 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5616 'video title', default=None)
5617 if title:
5618 title = title.replace('\xad', '') # remove soft hyphens
5619 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5620 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5621 else None)
5622 return {
5623 '_type': 'url',
5624 'url': url,
5625 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5626 'video_id': video_id,
5627 'title': title,
5628 'channel_id': channel_id,
5629 'channel': channel,
5630 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5631 'upload_date': upload_date,
5632 }
5633
5634 def _notification_menu_entries(self, ytcfg):
5635 continuation_list = [None]
5636 response = None
5637 for page in itertools.count(1):
5638 ctoken = traverse_obj(
5639 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5640 response = self._extract_response(
5641 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5642 ep='notification/get_notification_menu', check_get_keys='actions',
5643 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5644 yield from self._extract_notification_menu(response, continuation_list)
5645 if not continuation_list[0]:
5646 break
5647
5648 def _real_extract(self, url):
5649 display_id = 'notifications'
5650 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5651 self._report_playlist_authcheck(ytcfg)
5652 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5653
5654
5655 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5656 IE_DESC = 'YouTube search'
5657 IE_NAME = 'youtube:search'
5658 _SEARCH_KEY = 'ytsearch'
5659 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5660 _TESTS = [{
5661 'url': 'ytsearch5:youtube-dl test video',
5662 'playlist_count': 5,
5663 'info_dict': {
5664 'id': 'youtube-dl test video',
5665 'title': 'youtube-dl test video',
5666 }
5667 }]
5668
5669
5670 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5671 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5672 _SEARCH_KEY = 'ytsearchdate'
5673 IE_DESC = 'YouTube search, newest videos first'
5674 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5675 _TESTS = [{
5676 'url': 'ytsearchdate5:youtube-dl test video',
5677 'playlist_count': 5,
5678 'info_dict': {
5679 'id': 'youtube-dl test video',
5680 'title': 'youtube-dl test video',
5681 }
5682 }]
5683
5684
5685 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5686 IE_DESC = 'YouTube search URLs with sorting and filter support'
5687 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5688 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5689 _TESTS = [{
5690 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5691 'playlist_mincount': 5,
5692 'info_dict': {
5693 'id': 'youtube-dl test video',
5694 'title': 'youtube-dl test video',
5695 }
5696 }, {
5697 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5698 'playlist_mincount': 5,
5699 'info_dict': {
5700 'id': 'python',
5701 'title': 'python',
5702 }
5703 }, {
5704 'url': 'https://www.youtube.com/results?search_query=%23cats',
5705 'playlist_mincount': 1,
5706 'info_dict': {
5707 'id': '#cats',
5708 'title': '#cats',
5709 'entries': [{
5710 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5711 'title': '#cats',
5712 }],
5713 },
5714 }, {
5715 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5716 'only_matching': True,
5717 }]
5718
5719 def _real_extract(self, url):
5720 qs = parse_qs(url)
5721 query = (qs.get('search_query') or qs.get('q'))[0]
5722 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5723
5724
5725 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5726 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5727 IE_NAME = 'youtube:music:search_url'
5728 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5729 _TESTS = [{
5730 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5731 'playlist_count': 16,
5732 'info_dict': {
5733 'id': 'royalty free music',
5734 'title': 'royalty free music',
5735 }
5736 }, {
5737 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5738 'playlist_mincount': 30,
5739 'info_dict': {
5740 'id': 'royalty free music - songs',
5741 'title': 'royalty free music - songs',
5742 },
5743 'params': {'extract_flat': 'in_playlist'}
5744 }, {
5745 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5746 'playlist_mincount': 30,
5747 'info_dict': {
5748 'id': 'royalty free music - community playlists',
5749 'title': 'royalty free music - community playlists',
5750 },
5751 'params': {'extract_flat': 'in_playlist'}
5752 }]
5753
5754 _SECTIONS = {
5755 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5756 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5757 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5758 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5759 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5760 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5761 }
5762
5763 def _real_extract(self, url):
5764 qs = parse_qs(url)
5765 query = (qs.get('search_query') or qs.get('q'))[0]
5766 params = qs.get('sp', (None,))[0]
5767 if params:
5768 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5769 else:
5770 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5771 params = self._SECTIONS.get(section)
5772 if not params:
5773 section = None
5774 title = join_nonempty(query, section, delim=' - ')
5775 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5776
5777
5778 class YoutubeFeedsInfoExtractor(InfoExtractor):
5779 """
5780 Base class for feed extractors
5781 Subclasses must define the _FEED_NAME property.
5782 """
5783 _LOGIN_REQUIRED = True
5784
5785 def _real_initialize(self):
5786 YoutubeBaseInfoExtractor._check_login_required(self)
5787
5788 @property
5789 def IE_NAME(self):
5790 return 'youtube:%s' % self._FEED_NAME
5791
5792 def _real_extract(self, url):
5793 return self.url_result(
5794 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5795
5796
5797 class YoutubeWatchLaterIE(InfoExtractor):
5798 IE_NAME = 'youtube:watchlater'
5799 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5800 _VALID_URL = r':ytwatchlater'
5801 _TESTS = [{
5802 'url': ':ytwatchlater',
5803 'only_matching': True,
5804 }]
5805
5806 def _real_extract(self, url):
5807 return self.url_result(
5808 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5809
5810
5811 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5812 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5813 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5814 _FEED_NAME = 'recommended'
5815 _LOGIN_REQUIRED = False
5816 _TESTS = [{
5817 'url': ':ytrec',
5818 'only_matching': True,
5819 }, {
5820 'url': ':ytrecommended',
5821 'only_matching': True,
5822 }, {
5823 'url': 'https://youtube.com',
5824 'only_matching': True,
5825 }]
5826
5827
5828 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5829 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5830 _VALID_URL = r':ytsub(?:scription)?s?'
5831 _FEED_NAME = 'subscriptions'
5832 _TESTS = [{
5833 'url': ':ytsubs',
5834 'only_matching': True,
5835 }, {
5836 'url': ':ytsubscriptions',
5837 'only_matching': True,
5838 }]
5839
5840
5841 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5842 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5843 _VALID_URL = r':ythis(?:tory)?'
5844 _FEED_NAME = 'history'
5845 _TESTS = [{
5846 'url': ':ythistory',
5847 'only_matching': True,
5848 }]
5849
5850
5851 class YoutubeStoriesIE(InfoExtractor):
5852 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5853 IE_NAME = 'youtube:stories'
5854 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5855 _TESTS = [{
5856 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5857 'only_matching': True,
5858 }]
5859
5860 def _real_extract(self, url):
5861 playlist_id = f'RLTD{self._match_id(url)}'
5862 return self.url_result(
5863 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5864 ie=YoutubeTabIE, video_id=playlist_id)
5865
5866
5867 class YoutubeTruncatedURLIE(InfoExtractor):
5868 IE_NAME = 'youtube:truncated_url'
5869 IE_DESC = False # Do not list
5870 _VALID_URL = r'''(?x)
5871 (?:https?://)?
5872 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5873 (?:watch\?(?:
5874 feature=[a-z_]+|
5875 annotation_id=annotation_[^&]+|
5876 x-yt-cl=[0-9]+|
5877 hl=[^&]*|
5878 t=[0-9]+
5879 )?
5880 |
5881 attribution_link\?a=[^&]+
5882 )
5883 $
5884 '''
5885
5886 _TESTS = [{
5887 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5888 'only_matching': True,
5889 }, {
5890 'url': 'https://www.youtube.com/watch?',
5891 'only_matching': True,
5892 }, {
5893 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5894 'only_matching': True,
5895 }, {
5896 'url': 'https://www.youtube.com/watch?feature=foo',
5897 'only_matching': True,
5898 }, {
5899 'url': 'https://www.youtube.com/watch?hl=en-GB',
5900 'only_matching': True,
5901 }, {
5902 'url': 'https://www.youtube.com/watch?t=2372',
5903 'only_matching': True,
5904 }]
5905
5906 def _real_extract(self, url):
5907 raise ExtractorError(
5908 'Did you forget to quote the URL? Remember that & is a meta '
5909 'character in most shells, so you want to put the URL in quotes, '
5910 'like youtube-dl '
5911 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5912 ' or simply youtube-dl BaW_jenozKc .',
5913 expected=True)
5914
5915
5916 class YoutubeClipIE(InfoExtractor):
5917 IE_NAME = 'youtube:clip'
5918 IE_DESC = False # Do not list
5919 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5920
5921 def _real_extract(self, url):
5922 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5923 return self.url_result(url, 'Generic')
5924
5925
5926 class YoutubeTruncatedIDIE(InfoExtractor):
5927 IE_NAME = 'youtube:truncated_id'
5928 IE_DESC = False # Do not list
5929 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5930
5931 _TESTS = [{
5932 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5933 'only_matching': True,
5934 }]
5935
5936 def _real_extract(self, url):
5937 video_id = self._match_id(url)
5938 raise ExtractorError(
5939 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5940 expected=True)