]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[VideocampusSachsen] Improve extractor (#3604)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
... / ...
CommitLineData
1import base64
2import calendar
3import copy
4import datetime
5import functools
6import hashlib
7import itertools
8import json
9import math
10import os.path
11import random
12import re
13import sys
14import threading
15import time
16import traceback
17
18from .common import InfoExtractor, SearchInfoExtractor
19from ..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)
29from ..jsinterp import JSInterpreter
30from ..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
73INNERTUBE_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
232def _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
240def 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
268build_innertube_clients()
269
270
271class 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
892class 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 'source_preference': -10 if throttled else -1,
3187 'fps': int_or_none(fmt.get('fps')) or None,
3188 'height': height,
3189 'quality': q(quality),
3190 'has_drm': bool(fmt.get('drmFamilies')),
3191 'tbr': tbr,
3192 'url': fmt_url,
3193 'width': int_or_none(fmt.get('width')),
3194 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3195 'desc' if language_preference < -1 else ''),
3196 'language_preference': language_preference,
3197 # Strictly de-prioritize damaged and 3gp formats
3198 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3199 }
3200 mime_mobj = re.match(
3201 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3202 if mime_mobj:
3203 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3204 dct.update(parse_codecs(mime_mobj.group(2)))
3205 no_audio = dct.get('acodec') == 'none'
3206 no_video = dct.get('vcodec') == 'none'
3207 if no_audio:
3208 dct['vbr'] = tbr
3209 if no_video:
3210 dct['abr'] = tbr
3211 if no_audio or no_video:
3212 dct['downloader_options'] = {
3213 # Youtube throttles chunks >~10M
3214 'http_chunk_size': 10485760,
3215 }
3216 if dct.get('ext'):
3217 dct['container'] = dct['ext'] + '_dash'
3218 yield dct
3219
3220 live_from_start = is_live and self.get_param('live_from_start')
3221 skip_manifests = self._configuration_arg('skip')
3222 if not self.get_param('youtube_include_hls_manifest', True):
3223 skip_manifests.append('hls')
3224 get_dash = 'dash' not in skip_manifests and (
3225 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3226 get_hls = not live_from_start and 'hls' not in skip_manifests
3227
3228 def process_manifest_format(f, proto, itag):
3229 if itag in itags:
3230 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3231 return False
3232 itag = f'{itag}-{proto}'
3233 if itag:
3234 f['format_id'] = itag
3235 itags[itag] = proto
3236
3237 f['quality'] = next((
3238 q(qdict[val])
3239 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3240 if val in qdict), -1)
3241 return True
3242
3243 for sd in streaming_data:
3244 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3245 if hls_manifest_url:
3246 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3247 if process_manifest_format(f, 'hls', self._search_regex(
3248 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3249 yield f
3250
3251 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3252 if dash_manifest_url:
3253 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3254 if process_manifest_format(f, 'dash', f['format_id']):
3255 f['filesize'] = int_or_none(self._search_regex(
3256 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3257 if live_from_start:
3258 f['is_from_start'] = True
3259
3260 yield f
3261
3262 def _extract_storyboard(self, player_responses, duration):
3263 spec = get_first(
3264 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3265 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3266 if not base_url:
3267 return
3268 L = len(spec) - 1
3269 for i, args in enumerate(spec):
3270 args = args.split('#')
3271 counts = list(map(int_or_none, args[:5]))
3272 if len(args) != 8 or not all(counts):
3273 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3274 continue
3275 width, height, frame_count, cols, rows = counts
3276 N, sigh = args[6:]
3277
3278 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3279 fragment_count = frame_count / (cols * rows)
3280 fragment_duration = duration / fragment_count
3281 yield {
3282 'format_id': f'sb{i}',
3283 'format_note': 'storyboard',
3284 'ext': 'mhtml',
3285 'protocol': 'mhtml',
3286 'acodec': 'none',
3287 'vcodec': 'none',
3288 'url': url,
3289 'width': width,
3290 'height': height,
3291 'fragments': [{
3292 'url': url.replace('$M', str(j)),
3293 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3294 } for j in range(math.ceil(fragment_count))],
3295 }
3296
3297 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3298 webpage = None
3299 if 'webpage' not in self._configuration_arg('player_skip'):
3300 webpage = self._download_webpage(
3301 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3302
3303 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3304
3305 player_responses, player_url = self._extract_player_responses(
3306 self._get_requested_clients(url, smuggled_data),
3307 video_id, webpage, master_ytcfg)
3308
3309 return webpage, master_ytcfg, player_responses, player_url
3310
3311 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3312 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3313 is_live = get_first(video_details, 'isLive')
3314 if is_live is None:
3315 is_live = get_first(live_broadcast_details, 'isLiveNow')
3316
3317 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3318 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3319
3320 return live_broadcast_details, is_live, streaming_data, formats
3321
3322 def _real_extract(self, url):
3323 url, smuggled_data = unsmuggle_url(url, {})
3324 video_id = self._match_id(url)
3325
3326 base_url = self.http_scheme() + '//www.youtube.com/'
3327 webpage_url = base_url + 'watch?v=' + video_id
3328
3329 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3330
3331 playability_statuses = traverse_obj(
3332 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3333
3334 trailer_video_id = get_first(
3335 playability_statuses,
3336 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3337 expected_type=str)
3338 if trailer_video_id:
3339 return self.url_result(
3340 trailer_video_id, self.ie_key(), trailer_video_id)
3341
3342 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3343 if webpage else (lambda x: None))
3344
3345 video_details = traverse_obj(
3346 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3347 microformats = traverse_obj(
3348 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3349 expected_type=dict, default=[])
3350 video_title = (
3351 get_first(video_details, 'title')
3352 or self._get_text(microformats, (..., 'title'))
3353 or search_meta(['og:title', 'twitter:title', 'title']))
3354 video_description = get_first(video_details, 'shortDescription')
3355
3356 multifeed_metadata_list = get_first(
3357 player_responses,
3358 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3359 expected_type=str)
3360 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3361 if self.get_param('noplaylist'):
3362 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3363 else:
3364 entries = []
3365 feed_ids = []
3366 for feed in multifeed_metadata_list.split(','):
3367 # Unquote should take place before split on comma (,) since textual
3368 # fields may contain comma as well (see
3369 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3370 feed_data = compat_parse_qs(
3371 compat_urllib_parse_unquote_plus(feed))
3372
3373 def feed_entry(name):
3374 return try_get(
3375 feed_data, lambda x: x[name][0], compat_str)
3376
3377 feed_id = feed_entry('id')
3378 if not feed_id:
3379 continue
3380 feed_title = feed_entry('title')
3381 title = video_title
3382 if feed_title:
3383 title += ' (%s)' % feed_title
3384 entries.append({
3385 '_type': 'url_transparent',
3386 'ie_key': 'Youtube',
3387 'url': smuggle_url(
3388 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3389 {'force_singlefeed': True}),
3390 'title': title,
3391 })
3392 feed_ids.append(feed_id)
3393 self.to_screen(
3394 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3395 % (', '.join(feed_ids), video_id))
3396 return self.playlist_result(
3397 entries, video_id, video_title, video_description)
3398
3399 duration = int_or_none(
3400 get_first(video_details, 'lengthSeconds')
3401 or get_first(microformats, 'lengthSeconds')
3402 or parse_duration(search_meta('duration'))) or None
3403
3404 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3405 video_id, microformats, video_details, player_responses, player_url, duration)
3406
3407 if not formats:
3408 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3409 self.report_drm(video_id)
3410 pemr = get_first(
3411 playability_statuses,
3412 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3413 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3414 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3415 if subreason:
3416 if subreason == 'The uploader has not made this video available in your country.':
3417 countries = get_first(microformats, 'availableCountries')
3418 if not countries:
3419 regions_allowed = search_meta('regionsAllowed')
3420 countries = regions_allowed.split(',') if regions_allowed else None
3421 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3422 reason += f'. {subreason}'
3423 if reason:
3424 self.raise_no_formats(reason, expected=True)
3425
3426 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3427 if not keywords and webpage:
3428 keywords = [
3429 unescapeHTML(m.group('content'))
3430 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3431 for keyword in keywords:
3432 if keyword.startswith('yt:stretch='):
3433 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3434 if mobj:
3435 # NB: float is intentional for forcing float division
3436 w, h = (float(v) for v in mobj.groups())
3437 if w > 0 and h > 0:
3438 ratio = w / h
3439 for f in formats:
3440 if f.get('vcodec') != 'none':
3441 f['stretched_ratio'] = ratio
3442 break
3443 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3444 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3445 if thumbnail_url:
3446 thumbnails.append({
3447 'url': thumbnail_url,
3448 })
3449 original_thumbnails = thumbnails.copy()
3450
3451 # The best resolution thumbnails sometimes does not appear in the webpage
3452 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3453 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3454 thumbnail_names = [
3455 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3456 # in resolution, these are not the custom thumbnail. So de-prioritize them
3457 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3458 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3459 ]
3460 n_thumbnail_names = len(thumbnail_names)
3461 thumbnails.extend({
3462 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3463 video_id=video_id, name=name, ext=ext,
3464 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3465 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3466 for thumb in thumbnails:
3467 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3468 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3469 self._remove_duplicate_formats(thumbnails)
3470 self._downloader._sort_thumbnails(original_thumbnails)
3471
3472 category = get_first(microformats, 'category') or search_meta('genre')
3473 channel_id = str_or_none(
3474 get_first(video_details, 'channelId')
3475 or get_first(microformats, 'externalChannelId')
3476 or search_meta('channelId'))
3477 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3478
3479 live_content = get_first(video_details, 'isLiveContent')
3480 is_upcoming = get_first(video_details, 'isUpcoming')
3481 if is_live is None:
3482 if is_upcoming or live_content is False:
3483 is_live = False
3484 if is_upcoming is None and (live_content or is_live):
3485 is_upcoming = False
3486 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3487 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3488 if not duration and live_end_time and live_start_time:
3489 duration = live_end_time - live_start_time
3490
3491 if is_live and self.get_param('live_from_start'):
3492 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3493
3494 formats.extend(self._extract_storyboard(player_responses, duration))
3495
3496 # Source is given priority since formats that throttle are given lower source_preference
3497 # When throttling issue is fully fixed, remove this
3498 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3499
3500 info = {
3501 'id': video_id,
3502 'title': video_title,
3503 'formats': formats,
3504 'thumbnails': thumbnails,
3505 # The best thumbnail that we are sure exists. Prevents unnecessary
3506 # URL checking if user don't care about getting the best possible thumbnail
3507 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3508 'description': video_description,
3509 'uploader': get_first(video_details, 'author'),
3510 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3511 'uploader_url': owner_profile_url,
3512 'channel_id': channel_id,
3513 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3514 'duration': duration,
3515 'view_count': int_or_none(
3516 get_first((video_details, microformats), (..., 'viewCount'))
3517 or search_meta('interactionCount')),
3518 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3519 'age_limit': 18 if (
3520 get_first(microformats, 'isFamilySafe') is False
3521 or search_meta('isFamilyFriendly') == 'false'
3522 or search_meta('og:restrictions:age') == '18+') else 0,
3523 'webpage_url': webpage_url,
3524 'categories': [category] if category else None,
3525 'tags': keywords,
3526 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3527 'is_live': is_live,
3528 'was_live': (False if is_live or is_upcoming or live_content is False
3529 else None if is_live is None or is_upcoming is None
3530 else live_content),
3531 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3532 'release_timestamp': live_start_time,
3533 }
3534
3535 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3536 if pctr:
3537 def get_lang_code(track):
3538 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3539 or track.get('languageCode'))
3540
3541 # Converted into dicts to remove duplicates
3542 captions = {
3543 get_lang_code(sub): sub
3544 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3545 translation_languages = {
3546 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3547 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3548
3549 def process_language(container, base_url, lang_code, sub_name, query):
3550 lang_subs = container.setdefault(lang_code, [])
3551 for fmt in self._SUBTITLE_FORMATS:
3552 query.update({
3553 'fmt': fmt,
3554 })
3555 lang_subs.append({
3556 'ext': fmt,
3557 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3558 'name': sub_name,
3559 })
3560
3561 subtitles, automatic_captions = {}, {}
3562 for lang_code, caption_track in captions.items():
3563 base_url = caption_track.get('baseUrl')
3564 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3565 if not base_url:
3566 continue
3567 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3568 if caption_track.get('kind') != 'asr':
3569 if not lang_code:
3570 continue
3571 process_language(
3572 subtitles, base_url, lang_code, lang_name, {})
3573 if not caption_track.get('isTranslatable'):
3574 continue
3575 for trans_code, trans_name in translation_languages.items():
3576 if not trans_code:
3577 continue
3578 orig_trans_code = trans_code
3579 if caption_track.get('kind') != 'asr':
3580 if 'translated_subs' in self._configuration_arg('skip'):
3581 continue
3582 trans_code += f'-{lang_code}'
3583 trans_name += format_field(lang_name, template=' from %s')
3584 # Add an "-orig" label to the original language so that it can be distinguished.
3585 # The subs are returned without "-orig" as well for compatibility
3586 if lang_code == f'a-{orig_trans_code}':
3587 process_language(
3588 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3589 # Setting tlang=lang returns damaged subtitles.
3590 process_language(automatic_captions, base_url, trans_code, trans_name,
3591 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3592 info['automatic_captions'] = automatic_captions
3593 info['subtitles'] = subtitles
3594
3595 parsed_url = compat_urllib_parse_urlparse(url)
3596 for component in [parsed_url.fragment, parsed_url.query]:
3597 query = compat_parse_qs(component)
3598 for k, v in query.items():
3599 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3600 d_k += '_time'
3601 if d_k not in info and k in s_ks:
3602 info[d_k] = parse_duration(query[k][0])
3603
3604 # Youtube Music Auto-generated description
3605 if video_description:
3606 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)
3607 if mobj:
3608 release_year = mobj.group('release_year')
3609 release_date = mobj.group('release_date')
3610 if release_date:
3611 release_date = release_date.replace('-', '')
3612 if not release_year:
3613 release_year = release_date[:4]
3614 info.update({
3615 'album': mobj.group('album'.strip()),
3616 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3617 'track': mobj.group('track').strip(),
3618 'release_date': release_date,
3619 'release_year': int_or_none(release_year),
3620 })
3621
3622 initial_data = None
3623 if webpage:
3624 initial_data = self._extract_yt_initial_variable(
3625 webpage, self._YT_INITIAL_DATA_RE, video_id,
3626 'yt initial data')
3627 if not initial_data:
3628 query = {'videoId': video_id}
3629 query.update(self._get_checkok_params())
3630 initial_data = self._extract_response(
3631 item_id=video_id, ep='next', fatal=False,
3632 ytcfg=master_ytcfg, query=query,
3633 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3634 note='Downloading initial data API JSON')
3635
3636 try: # This will error if there is no livechat
3637 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3638 except (KeyError, IndexError, TypeError):
3639 pass
3640 else:
3641 info.setdefault('subtitles', {})['live_chat'] = [{
3642 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3643 'video_id': video_id,
3644 'ext': 'json',
3645 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3646 }]
3647
3648 if initial_data:
3649 info['chapters'] = (
3650 self._extract_chapters_from_json(initial_data, duration)
3651 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3652 or None)
3653
3654 contents = traverse_obj(
3655 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3656 expected_type=list, default=[])
3657
3658 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3659 if vpir:
3660 stl = vpir.get('superTitleLink')
3661 if stl:
3662 stl = self._get_text(stl)
3663 if try_get(
3664 vpir,
3665 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3666 info['location'] = stl
3667 else:
3668 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3669 if mobj:
3670 info.update({
3671 'series': mobj.group(1),
3672 'season_number': int(mobj.group(2)),
3673 'episode_number': int(mobj.group(3)),
3674 })
3675 for tlb in (try_get(
3676 vpir,
3677 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3678 list) or []):
3679 tbr = tlb.get('toggleButtonRenderer') or {}
3680 for getter, regex in [(
3681 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3682 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3683 lambda x: x['accessibility'],
3684 lambda x: x['accessibilityData']['accessibilityData'],
3685 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3686 label = (try_get(tbr, getter, dict) or {}).get('label')
3687 if label:
3688 mobj = re.match(regex, label)
3689 if mobj:
3690 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3691 break
3692 sbr_tooltip = try_get(
3693 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3694 if sbr_tooltip:
3695 like_count, dislike_count = sbr_tooltip.split(' / ')
3696 info.update({
3697 'like_count': str_to_int(like_count),
3698 'dislike_count': str_to_int(dislike_count),
3699 })
3700 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3701 if vsir:
3702 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3703 info.update({
3704 'channel': self._get_text(vor, 'title'),
3705 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3706
3707 rows = try_get(
3708 vsir,
3709 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3710 list) or []
3711 multiple_songs = False
3712 for row in rows:
3713 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3714 multiple_songs = True
3715 break
3716 for row in rows:
3717 mrr = row.get('metadataRowRenderer') or {}
3718 mrr_title = mrr.get('title')
3719 if not mrr_title:
3720 continue
3721 mrr_title = self._get_text(mrr, 'title')
3722 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3723 if mrr_title == 'License':
3724 info['license'] = mrr_contents_text
3725 elif not multiple_songs:
3726 if mrr_title == 'Album':
3727 info['album'] = mrr_contents_text
3728 elif mrr_title == 'Artist':
3729 info['artist'] = mrr_contents_text
3730 elif mrr_title == 'Song':
3731 info['track'] = mrr_contents_text
3732
3733 fallbacks = {
3734 'channel': 'uploader',
3735 'channel_id': 'uploader_id',
3736 'channel_url': 'uploader_url',
3737 }
3738
3739 # The upload date for scheduled, live and past live streams / premieres in microformats
3740 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3741 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3742 upload_date = (
3743 unified_strdate(get_first(microformats, 'uploadDate'))
3744 or unified_strdate(search_meta('uploadDate')))
3745 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3746 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3747 info['upload_date'] = upload_date
3748
3749 for to, frm in fallbacks.items():
3750 if not info.get(to):
3751 info[to] = info.get(frm)
3752
3753 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3754 v = info.get(s_k)
3755 if v:
3756 info[d_k] = v
3757
3758 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3759 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3760 is_membersonly = None
3761 is_premium = None
3762 if initial_data and is_private is not None:
3763 is_membersonly = False
3764 is_premium = False
3765 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3766 badge_labels = set()
3767 for content in contents:
3768 if not isinstance(content, dict):
3769 continue
3770 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3771 for badge_label in badge_labels:
3772 if badge_label.lower() == 'members only':
3773 is_membersonly = True
3774 elif badge_label.lower() == 'premium':
3775 is_premium = True
3776 elif badge_label.lower() == 'unlisted':
3777 is_unlisted = True
3778
3779 info['availability'] = self._availability(
3780 is_private=is_private,
3781 needs_premium=is_premium,
3782 needs_subscription=is_membersonly,
3783 needs_auth=info['age_limit'] >= 18,
3784 is_unlisted=None if is_private is None else is_unlisted)
3785
3786 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3787
3788 self.mark_watched(video_id, player_responses)
3789
3790 return info
3791
3792
3793class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3794
3795 @staticmethod
3796 def passthrough_smuggled_data(func):
3797 def _smuggle(entries, smuggled_data):
3798 for entry in entries:
3799 # TODO: Convert URL to music.youtube instead.
3800 # Do we need to passthrough any other smuggled_data?
3801 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3802 yield entry
3803
3804 @functools.wraps(func)
3805 def wrapper(self, url):
3806 url, smuggled_data = unsmuggle_url(url, {})
3807 if self.is_music_url(url):
3808 smuggled_data['is_music_url'] = True
3809 info_dict = func(self, url, smuggled_data)
3810 if smuggled_data and info_dict.get('entries'):
3811 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3812 return info_dict
3813 return wrapper
3814
3815 def _extract_channel_id(self, webpage):
3816 channel_id = self._html_search_meta(
3817 'channelId', webpage, 'channel id', default=None)
3818 if channel_id:
3819 return channel_id
3820 channel_url = self._html_search_meta(
3821 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3822 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3823 'twitter:app:url:googleplay'), webpage, 'channel url')
3824 return self._search_regex(
3825 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3826 channel_url, 'channel id')
3827
3828 @staticmethod
3829 def _extract_basic_item_renderer(item):
3830 # Modified from _extract_grid_item_renderer
3831 known_basic_renderers = (
3832 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3833 )
3834 for key, renderer in item.items():
3835 if not isinstance(renderer, dict):
3836 continue
3837 elif key in known_basic_renderers:
3838 return renderer
3839 elif key.startswith('grid') and key.endswith('Renderer'):
3840 return renderer
3841
3842 def _grid_entries(self, grid_renderer):
3843 for item in grid_renderer['items']:
3844 if not isinstance(item, dict):
3845 continue
3846 renderer = self._extract_basic_item_renderer(item)
3847 if not isinstance(renderer, dict):
3848 continue
3849 title = self._get_text(renderer, 'title')
3850
3851 # playlist
3852 playlist_id = renderer.get('playlistId')
3853 if playlist_id:
3854 yield self.url_result(
3855 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3856 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3857 video_title=title)
3858 continue
3859 # video
3860 video_id = renderer.get('videoId')
3861 if video_id:
3862 yield self._extract_video(renderer)
3863 continue
3864 # channel
3865 channel_id = renderer.get('channelId')
3866 if channel_id:
3867 yield self.url_result(
3868 'https://www.youtube.com/channel/%s' % channel_id,
3869 ie=YoutubeTabIE.ie_key(), video_title=title)
3870 continue
3871 # generic endpoint URL support
3872 ep_url = urljoin('https://www.youtube.com/', try_get(
3873 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3874 compat_str))
3875 if ep_url:
3876 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3877 if ie.suitable(ep_url):
3878 yield self.url_result(
3879 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3880 break
3881
3882 def _music_reponsive_list_entry(self, renderer):
3883 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3884 if video_id:
3885 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3886 ie=YoutubeIE.ie_key(), video_id=video_id)
3887 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3888 if playlist_id:
3889 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3890 if video_id:
3891 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3892 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3893 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3894 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3895 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3896 if browse_id:
3897 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3898 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3899
3900 def _shelf_entries_from_content(self, shelf_renderer):
3901 content = shelf_renderer.get('content')
3902 if not isinstance(content, dict):
3903 return
3904 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3905 if renderer:
3906 # TODO: add support for nested playlists so each shelf is processed
3907 # as separate playlist
3908 # TODO: this includes only first N items
3909 yield from self._grid_entries(renderer)
3910 renderer = content.get('horizontalListRenderer')
3911 if renderer:
3912 # TODO
3913 pass
3914
3915 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3916 ep = try_get(
3917 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3918 compat_str)
3919 shelf_url = urljoin('https://www.youtube.com', ep)
3920 if shelf_url:
3921 # Skipping links to another channels, note that checking for
3922 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3923 # will not work
3924 if skip_channels and '/channels?' in shelf_url:
3925 return
3926 title = self._get_text(shelf_renderer, 'title')
3927 yield self.url_result(shelf_url, video_title=title)
3928 # Shelf may not contain shelf URL, fallback to extraction from content
3929 yield from self._shelf_entries_from_content(shelf_renderer)
3930
3931 def _playlist_entries(self, video_list_renderer):
3932 for content in video_list_renderer['contents']:
3933 if not isinstance(content, dict):
3934 continue
3935 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3936 if not isinstance(renderer, dict):
3937 continue
3938 video_id = renderer.get('videoId')
3939 if not video_id:
3940 continue
3941 yield self._extract_video(renderer)
3942
3943 def _rich_entries(self, rich_grid_renderer):
3944 renderer = try_get(
3945 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3946 video_id = renderer.get('videoId')
3947 if not video_id:
3948 return
3949 yield self._extract_video(renderer)
3950
3951 def _video_entry(self, video_renderer):
3952 video_id = video_renderer.get('videoId')
3953 if video_id:
3954 return self._extract_video(video_renderer)
3955
3956 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3957 url = urljoin('https://youtube.com', traverse_obj(
3958 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3959 if url:
3960 return self.url_result(
3961 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3962
3963 def _post_thread_entries(self, post_thread_renderer):
3964 post_renderer = try_get(
3965 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3966 if not post_renderer:
3967 return
3968 # video attachment
3969 video_renderer = try_get(
3970 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3971 video_id = video_renderer.get('videoId')
3972 if video_id:
3973 entry = self._extract_video(video_renderer)
3974 if entry:
3975 yield entry
3976 # playlist attachment
3977 playlist_id = try_get(
3978 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3979 if playlist_id:
3980 yield self.url_result(
3981 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3982 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3983 # inline video links
3984 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
3985 for run in runs:
3986 if not isinstance(run, dict):
3987 continue
3988 ep_url = try_get(
3989 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
3990 if not ep_url:
3991 continue
3992 if not YoutubeIE.suitable(ep_url):
3993 continue
3994 ep_video_id = YoutubeIE._match_id(ep_url)
3995 if video_id == ep_video_id:
3996 continue
3997 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
3998
3999 def _post_thread_continuation_entries(self, post_thread_continuation):
4000 contents = post_thread_continuation.get('contents')
4001 if not isinstance(contents, list):
4002 return
4003 for content in contents:
4004 renderer = content.get('backstagePostThreadRenderer')
4005 if not isinstance(renderer, dict):
4006 continue
4007 yield from self._post_thread_entries(renderer)
4008
4009 r''' # unused
4010 def _rich_grid_entries(self, contents):
4011 for content in contents:
4012 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4013 if video_renderer:
4014 entry = self._video_entry(video_renderer)
4015 if entry:
4016 yield entry
4017 '''
4018
4019 def _extract_entries(self, parent_renderer, continuation_list):
4020 # continuation_list is modified in-place with continuation_list = [continuation_token]
4021 continuation_list[:] = [None]
4022 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4023 for content in contents:
4024 if not isinstance(content, dict):
4025 continue
4026 is_renderer = traverse_obj(
4027 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4028 expected_type=dict)
4029 if not is_renderer:
4030 renderer = content.get('richItemRenderer')
4031 if renderer:
4032 for entry in self._rich_entries(renderer):
4033 yield entry
4034 continuation_list[0] = self._extract_continuation(parent_renderer)
4035 continue
4036 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4037 for isr_content in isr_contents:
4038 if not isinstance(isr_content, dict):
4039 continue
4040
4041 known_renderers = {
4042 'playlistVideoListRenderer': self._playlist_entries,
4043 'gridRenderer': self._grid_entries,
4044 'reelShelfRenderer': self._grid_entries,
4045 'shelfRenderer': self._shelf_entries,
4046 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4047 'backstagePostThreadRenderer': self._post_thread_entries,
4048 'videoRenderer': lambda x: [self._video_entry(x)],
4049 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4050 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4051 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4052 }
4053 for key, renderer in isr_content.items():
4054 if key not in known_renderers:
4055 continue
4056 for entry in known_renderers[key](renderer):
4057 if entry:
4058 yield entry
4059 continuation_list[0] = self._extract_continuation(renderer)
4060 break
4061
4062 if not continuation_list[0]:
4063 continuation_list[0] = self._extract_continuation(is_renderer)
4064
4065 if not continuation_list[0]:
4066 continuation_list[0] = self._extract_continuation(parent_renderer)
4067
4068 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4069 continuation_list = [None]
4070 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4071 tab_content = try_get(tab, lambda x: x['content'], dict)
4072 if not tab_content:
4073 return
4074 parent_renderer = (
4075 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4076 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4077 yield from extract_entries(parent_renderer)
4078 continuation = continuation_list[0]
4079
4080 for page_num in itertools.count(1):
4081 if not continuation:
4082 break
4083 headers = self.generate_api_headers(
4084 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4085 response = self._extract_response(
4086 item_id=f'{item_id} page {page_num}',
4087 query=continuation, headers=headers, ytcfg=ytcfg,
4088 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4089
4090 if not response:
4091 break
4092 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4093 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4094 visitor_data = self._extract_visitor_data(response) or visitor_data
4095
4096 known_continuation_renderers = {
4097 'playlistVideoListContinuation': self._playlist_entries,
4098 'gridContinuation': self._grid_entries,
4099 'itemSectionContinuation': self._post_thread_continuation_entries,
4100 'sectionListContinuation': extract_entries, # for feeds
4101 }
4102 continuation_contents = try_get(
4103 response, lambda x: x['continuationContents'], dict) or {}
4104 continuation_renderer = None
4105 for key, value in continuation_contents.items():
4106 if key not in known_continuation_renderers:
4107 continue
4108 continuation_renderer = value
4109 continuation_list = [None]
4110 yield from known_continuation_renderers[key](continuation_renderer)
4111 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4112 break
4113 if continuation_renderer:
4114 continue
4115
4116 known_renderers = {
4117 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4118 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4119 'gridVideoRenderer': (self._grid_entries, 'items'),
4120 'gridChannelRenderer': (self._grid_entries, 'items'),
4121 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4122 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4123 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4124 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4125 }
4126 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4127 continuation_items = try_get(
4128 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4129 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4130 video_items_renderer = None
4131 for key, value in continuation_item.items():
4132 if key not in known_renderers:
4133 continue
4134 video_items_renderer = {known_renderers[key][1]: continuation_items}
4135 continuation_list = [None]
4136 yield from known_renderers[key][0](video_items_renderer)
4137 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4138 break
4139 if video_items_renderer:
4140 continue
4141 break
4142
4143 @staticmethod
4144 def _extract_selected_tab(tabs, fatal=True):
4145 for tab in tabs:
4146 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4147 if renderer.get('selected') is True:
4148 return renderer
4149 else:
4150 if fatal:
4151 raise ExtractorError('Unable to find selected tab')
4152
4153 def _extract_uploader(self, data):
4154 uploader = {}
4155 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4156 owner = try_get(
4157 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4158 if owner:
4159 owner_text = owner.get('text')
4160 uploader['uploader'] = self._search_regex(
4161 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4162 uploader['uploader_id'] = try_get(
4163 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4164 uploader['uploader_url'] = urljoin(
4165 'https://www.youtube.com/',
4166 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4167 return {k: v for k, v in uploader.items() if v is not None}
4168
4169 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4170 playlist_id = title = description = channel_url = channel_name = channel_id = None
4171 tags = []
4172
4173 selected_tab = self._extract_selected_tab(tabs)
4174 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4175 renderer = try_get(
4176 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4177 if renderer:
4178 channel_name = renderer.get('title')
4179 channel_url = renderer.get('channelUrl')
4180 channel_id = renderer.get('externalId')
4181 else:
4182 renderer = try_get(
4183 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4184
4185 if renderer:
4186 title = renderer.get('title')
4187 description = renderer.get('description', '')
4188 playlist_id = channel_id
4189 tags = renderer.get('keywords', '').split()
4190
4191 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4192 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4193 def _get_uncropped(url):
4194 return url_or_none((url or '').split('=')[0] + '=s0')
4195
4196 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4197 if avatar_thumbnails:
4198 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4199 if uncropped_avatar:
4200 avatar_thumbnails.append({
4201 'url': uncropped_avatar,
4202 'id': 'avatar_uncropped',
4203 'preference': 1
4204 })
4205
4206 channel_banners = self._extract_thumbnails(
4207 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4208 for banner in channel_banners:
4209 banner['preference'] = -10
4210
4211 if channel_banners:
4212 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4213 if uncropped_banner:
4214 channel_banners.append({
4215 'url': uncropped_banner,
4216 'id': 'banner_uncropped',
4217 'preference': -5
4218 })
4219
4220 primary_thumbnails = self._extract_thumbnails(
4221 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4222
4223 if playlist_id is None:
4224 playlist_id = item_id
4225
4226 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4227 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4228 if title is None:
4229 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4230 title += format_field(selected_tab, 'title', ' - %s')
4231 title += format_field(selected_tab, 'expandedText', ' - %s')
4232
4233 metadata = {
4234 'playlist_id': playlist_id,
4235 'playlist_title': title,
4236 'playlist_description': description,
4237 'uploader': channel_name,
4238 'uploader_id': channel_id,
4239 'uploader_url': channel_url,
4240 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4241 'tags': tags,
4242 'view_count': self._get_count(playlist_stats, 1),
4243 'availability': self._extract_availability(data),
4244 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4245 'playlist_count': self._get_count(playlist_stats, 0),
4246 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4247 }
4248 if not channel_id:
4249 metadata.update(self._extract_uploader(data))
4250 metadata.update({
4251 'channel': metadata['uploader'],
4252 'channel_id': metadata['uploader_id'],
4253 'channel_url': metadata['uploader_url']})
4254 return self.playlist_result(
4255 self._entries(
4256 selected_tab, playlist_id, ytcfg,
4257 self._extract_account_syncid(ytcfg, data),
4258 self._extract_visitor_data(data, ytcfg)),
4259 **metadata)
4260
4261 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4262 first_id = last_id = response = None
4263 for page_num in itertools.count(1):
4264 videos = list(self._playlist_entries(playlist))
4265 if not videos:
4266 return
4267 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4268 if start >= len(videos):
4269 return
4270 for video in videos[start:]:
4271 yield video
4272 first_id = first_id or videos[0]['id']
4273 last_id = videos[-1]['id']
4274 watch_endpoint = try_get(
4275 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4276 headers = self.generate_api_headers(
4277 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4278 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4279 query = {
4280 'playlistId': playlist_id,
4281 'videoId': watch_endpoint.get('videoId') or last_id,
4282 'index': watch_endpoint.get('index') or len(videos),
4283 'params': watch_endpoint.get('params') or 'OAE%3D'
4284 }
4285 response = self._extract_response(
4286 item_id='%s page %d' % (playlist_id, page_num),
4287 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4288 check_get_keys='contents'
4289 )
4290 playlist = try_get(
4291 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4292
4293 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4294 title = playlist.get('title') or try_get(
4295 data, lambda x: x['titleText']['simpleText'], compat_str)
4296 playlist_id = playlist.get('playlistId') or item_id
4297
4298 # Delegating everything except mix playlists to regular tab-based playlist URL
4299 playlist_url = urljoin(url, try_get(
4300 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4301 compat_str))
4302
4303 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4304 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4305 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4306
4307 if playlist_url and playlist_url != url and not is_known_unviewable:
4308 return self.url_result(
4309 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4310 video_title=title)
4311
4312 return self.playlist_result(
4313 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4314 playlist_id=playlist_id, playlist_title=title)
4315
4316 def _extract_availability(self, data):
4317 """
4318 Gets the availability of a given playlist/tab.
4319 Note: Unless YouTube tells us explicitly, we do not assume it is public
4320 @param data: response
4321 """
4322 is_private = is_unlisted = None
4323 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4324 badge_labels = self._extract_badges(renderer)
4325
4326 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4327 privacy_dropdown_entries = try_get(
4328 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4329 for renderer_dict in privacy_dropdown_entries:
4330 is_selected = try_get(
4331 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4332 if not is_selected:
4333 continue
4334 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4335 if label:
4336 badge_labels.add(label.lower())
4337 break
4338
4339 for badge_label in badge_labels:
4340 if badge_label == 'unlisted':
4341 is_unlisted = True
4342 elif badge_label == 'private':
4343 is_private = True
4344 elif badge_label == 'public':
4345 is_unlisted = is_private = False
4346 return self._availability(is_private, False, False, False, is_unlisted)
4347
4348 @staticmethod
4349 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4350 sidebar_renderer = try_get(
4351 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4352 for item in sidebar_renderer:
4353 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4354 if renderer:
4355 return renderer
4356
4357 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4358 """
4359 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4360 """
4361 browse_id = params = None
4362 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4363 if not renderer:
4364 return
4365 menu_renderer = try_get(
4366 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4367 for menu_item in menu_renderer:
4368 if not isinstance(menu_item, dict):
4369 continue
4370 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4371 text = try_get(
4372 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4373 if not text or text.lower() != 'show unavailable videos':
4374 continue
4375 browse_endpoint = try_get(
4376 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4377 browse_id = browse_endpoint.get('browseId')
4378 params = browse_endpoint.get('params')
4379 break
4380
4381 headers = self.generate_api_headers(
4382 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4383 visitor_data=self._extract_visitor_data(data, ytcfg))
4384 query = {
4385 'params': params or 'wgYCCAA=',
4386 'browseId': browse_id or 'VL%s' % item_id
4387 }
4388 return self._extract_response(
4389 item_id=item_id, headers=headers, query=query,
4390 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4391 note='Downloading API JSON with unavailable videos')
4392
4393 @property
4394 def skip_webpage(self):
4395 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4396
4397 def _extract_webpage(self, url, item_id, fatal=True):
4398 retries = self.get_param('extractor_retries', 3)
4399 count = -1
4400 webpage = data = last_error = None
4401 while count < retries:
4402 count += 1
4403 # Sometimes youtube returns a webpage with incomplete ytInitialData
4404 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4405 if last_error:
4406 self.report_warning('%s. Retrying ...' % last_error)
4407 try:
4408 webpage = self._download_webpage(
4409 url, item_id,
4410 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4411 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4412 except ExtractorError as e:
4413 if isinstance(e.cause, network_exceptions):
4414 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4415 last_error = error_to_compat_str(e.cause or e.msg)
4416 if count < retries:
4417 continue
4418 if fatal:
4419 raise
4420 self.report_warning(error_to_compat_str(e))
4421 break
4422 else:
4423 try:
4424 self._extract_and_report_alerts(data)
4425 except ExtractorError as e:
4426 if fatal:
4427 raise
4428 self.report_warning(error_to_compat_str(e))
4429 break
4430
4431 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4432 break
4433
4434 last_error = 'Incomplete yt initial data received'
4435 if count >= retries:
4436 if fatal:
4437 raise ExtractorError(last_error)
4438 self.report_warning(last_error)
4439 break
4440
4441 return webpage, data
4442
4443 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4444 """Use if failed to extract ytcfg (and data) from initial webpage"""
4445 if not ytcfg and self.is_authenticated:
4446 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4447 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4448 raise ExtractorError(
4449 f'{msg}. If you are not downloading private content, or '
4450 'your cookies are only for the first account and channel,'
4451 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4452 expected=True)
4453 self.report_warning(msg, only_once=True)
4454
4455 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4456 data = None
4457 if not self.skip_webpage:
4458 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4459 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4460 # Reject webpage data if redirected to home page without explicitly requesting
4461 selected_tab = self._extract_selected_tab(traverse_obj(
4462 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4463 if (url != 'https://www.youtube.com/feed/recommended'
4464 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4465 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4466 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4467 if fatal:
4468 raise ExtractorError(msg, expected=True)
4469 self.report_warning(msg, only_once=True)
4470 if not data:
4471 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4472 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4473 return data, ytcfg
4474
4475 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4476 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4477 resolve_response = self._extract_response(
4478 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4479 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4480 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4481 for ep_key, ep in endpoints.items():
4482 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4483 if params:
4484 return self._extract_response(
4485 item_id=item_id, query=params, ep=ep, headers=headers,
4486 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4487 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4488 err_note = 'Failed to resolve url (does the playlist exist?)'
4489 if fatal:
4490 raise ExtractorError(err_note, expected=True)
4491 self.report_warning(err_note, item_id)
4492
4493 _SEARCH_PARAMS = None
4494
4495 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4496 data = {'query': query}
4497 if params is NO_DEFAULT:
4498 params = self._SEARCH_PARAMS
4499 if params:
4500 data['params'] = params
4501
4502 content_keys = (
4503 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4504 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4505 # ytmusic search
4506 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4507 ('continuationContents', ),
4508 )
4509 display_id = f'query "{query}"'
4510 check_get_keys = tuple({keys[0] for keys in content_keys})
4511 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4512 self._report_playlist_authcheck(ytcfg, fatal=False)
4513
4514 continuation_list = [None]
4515 search = None
4516 for page_num in itertools.count(1):
4517 data.update(continuation_list[0] or {})
4518 headers = self.generate_api_headers(
4519 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4520 search = self._extract_response(
4521 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4522 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4523 slr_contents = traverse_obj(search, *content_keys)
4524 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4525 if not continuation_list[0]:
4526 break
4527
4528
4529class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4530 IE_DESC = 'YouTube Tabs'
4531 _VALID_URL = r'''(?x:
4532 https?://
4533 (?:\w+\.)?
4534 (?:
4535 youtube(?:kids)?\.com|
4536 %(invidious)s
4537 )/
4538 (?:
4539 (?P<channel_type>channel|c|user|browse)/|
4540 (?P<not_channel>
4541 feed/|hashtag/|
4542 (?:playlist|watch)\?.*?\blist=
4543 )|
4544 (?!(?:%(reserved_names)s)\b) # Direct URLs
4545 )
4546 (?P<id>[^/?\#&]+)
4547 )''' % {
4548 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4549 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4550 }
4551 IE_NAME = 'youtube:tab'
4552
4553 _TESTS = [{
4554 'note': 'playlists, multipage',
4555 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4556 'playlist_mincount': 94,
4557 'info_dict': {
4558 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4559 'title': 'Igor Kleiner - Playlists',
4560 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4561 'uploader': 'Igor Kleiner',
4562 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4563 'channel': 'Igor Kleiner',
4564 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4565 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4566 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4567 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4568 'channel_follower_count': int
4569 },
4570 }, {
4571 'note': 'playlists, multipage, different order',
4572 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4573 'playlist_mincount': 94,
4574 'info_dict': {
4575 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4576 'title': 'Igor Kleiner - Playlists',
4577 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4578 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4579 'uploader': 'Igor Kleiner',
4580 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4581 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4582 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4583 'channel': 'Igor Kleiner',
4584 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4585 'channel_follower_count': int
4586 },
4587 }, {
4588 'note': 'playlists, series',
4589 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4590 'playlist_mincount': 5,
4591 'info_dict': {
4592 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4593 'title': '3Blue1Brown - Playlists',
4594 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4595 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4596 'uploader': '3Blue1Brown',
4597 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4598 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4599 'channel': '3Blue1Brown',
4600 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4601 'tags': ['Mathematics'],
4602 'channel_follower_count': int
4603 },
4604 }, {
4605 'note': 'playlists, singlepage',
4606 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4607 'playlist_mincount': 4,
4608 'info_dict': {
4609 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4610 'title': 'ThirstForScience - Playlists',
4611 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4612 'uploader': 'ThirstForScience',
4613 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4614 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4615 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4616 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4617 'tags': 'count:13',
4618 'channel': 'ThirstForScience',
4619 'channel_follower_count': int
4620 }
4621 }, {
4622 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4623 'only_matching': True,
4624 }, {
4625 'note': 'basic, single video playlist',
4626 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4627 'info_dict': {
4628 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4629 'uploader': 'Sergey M.',
4630 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4631 'title': 'youtube-dl public playlist',
4632 'description': '',
4633 'tags': [],
4634 'view_count': int,
4635 'modified_date': '20201130',
4636 'channel': 'Sergey M.',
4637 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4638 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4639 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4640 },
4641 'playlist_count': 1,
4642 }, {
4643 'note': 'empty playlist',
4644 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4645 'info_dict': {
4646 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4647 'uploader': 'Sergey M.',
4648 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4649 'title': 'youtube-dl empty playlist',
4650 'tags': [],
4651 'channel': 'Sergey M.',
4652 'description': '',
4653 'modified_date': '20160902',
4654 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4655 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4656 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4657 },
4658 'playlist_count': 0,
4659 }, {
4660 'note': 'Home tab',
4661 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4662 'info_dict': {
4663 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4664 'title': 'lex will - Home',
4665 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4666 'uploader': 'lex will',
4667 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4668 'channel': 'lex will',
4669 'tags': ['bible', 'history', 'prophesy'],
4670 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4671 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4672 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4673 'channel_follower_count': int
4674 },
4675 'playlist_mincount': 2,
4676 }, {
4677 'note': 'Videos tab',
4678 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4679 'info_dict': {
4680 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4681 'title': 'lex will - Videos',
4682 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4683 'uploader': 'lex will',
4684 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4685 'tags': ['bible', 'history', 'prophesy'],
4686 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4687 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4688 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4689 'channel': 'lex will',
4690 'channel_follower_count': int
4691 },
4692 'playlist_mincount': 975,
4693 }, {
4694 'note': 'Videos tab, sorted by popular',
4695 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4696 'info_dict': {
4697 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4698 'title': 'lex will - Videos',
4699 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4700 'uploader': 'lex will',
4701 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4702 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4703 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4704 'channel': 'lex will',
4705 'tags': ['bible', 'history', 'prophesy'],
4706 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4707 'channel_follower_count': int
4708 },
4709 'playlist_mincount': 199,
4710 }, {
4711 'note': 'Playlists tab',
4712 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4713 'info_dict': {
4714 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4715 'title': 'lex will - Playlists',
4716 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4717 'uploader': 'lex will',
4718 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4719 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4720 'channel': 'lex will',
4721 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4722 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4723 'tags': ['bible', 'history', 'prophesy'],
4724 'channel_follower_count': int
4725 },
4726 'playlist_mincount': 17,
4727 }, {
4728 'note': 'Community tab',
4729 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4730 'info_dict': {
4731 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4732 'title': 'lex will - Community',
4733 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4734 'uploader': 'lex will',
4735 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4736 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4737 'channel': 'lex will',
4738 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4739 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4740 'tags': ['bible', 'history', 'prophesy'],
4741 'channel_follower_count': int
4742 },
4743 'playlist_mincount': 18,
4744 }, {
4745 'note': 'Channels tab',
4746 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4747 'info_dict': {
4748 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4749 'title': 'lex will - Channels',
4750 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4751 'uploader': 'lex will',
4752 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4753 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4754 'channel': 'lex will',
4755 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4756 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4757 'tags': ['bible', 'history', 'prophesy'],
4758 'channel_follower_count': int
4759 },
4760 'playlist_mincount': 12,
4761 }, {
4762 'note': 'Search tab',
4763 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4764 'playlist_mincount': 40,
4765 'info_dict': {
4766 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4767 'title': '3Blue1Brown - Search - linear algebra',
4768 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4769 'uploader': '3Blue1Brown',
4770 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4771 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4772 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4773 'tags': ['Mathematics'],
4774 'channel': '3Blue1Brown',
4775 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4776 'channel_follower_count': int
4777 },
4778 }, {
4779 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4780 'only_matching': True,
4781 }, {
4782 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4783 'only_matching': True,
4784 }, {
4785 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4786 'only_matching': True,
4787 }, {
4788 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4789 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4790 'info_dict': {
4791 'title': '29C3: Not my department',
4792 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4793 'uploader': 'Christiaan008',
4794 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4795 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4796 'tags': [],
4797 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4798 'view_count': int,
4799 'modified_date': '20150605',
4800 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4801 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4802 'channel': 'Christiaan008',
4803 },
4804 'playlist_count': 96,
4805 }, {
4806 'note': 'Large playlist',
4807 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4808 'info_dict': {
4809 'title': 'Uploads from Cauchemar',
4810 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4811 'uploader': 'Cauchemar',
4812 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4813 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4814 'tags': [],
4815 'modified_date': r're:\d{8}',
4816 'channel': 'Cauchemar',
4817 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4818 'view_count': int,
4819 'description': '',
4820 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4821 },
4822 'playlist_mincount': 1123,
4823 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4824 }, {
4825 'note': 'even larger playlist, 8832 videos',
4826 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4827 'only_matching': True,
4828 }, {
4829 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4830 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4831 'info_dict': {
4832 'title': 'Uploads from Interstellar Movie',
4833 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4834 'uploader': 'Interstellar Movie',
4835 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4836 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4837 'tags': [],
4838 'view_count': int,
4839 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4840 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4841 'channel': 'Interstellar Movie',
4842 'description': '',
4843 'modified_date': r're:\d{8}',
4844 },
4845 'playlist_mincount': 21,
4846 }, {
4847 'note': 'Playlist with "show unavailable videos" button',
4848 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4849 'info_dict': {
4850 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4851 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4852 'uploader': 'Phim Siêu Nhân Nhật Bản',
4853 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4854 'view_count': int,
4855 'channel': 'Phim Siêu Nhân Nhật Bản',
4856 'tags': [],
4857 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4858 'description': '',
4859 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4860 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4861 'modified_date': r're:\d{8}',
4862 },
4863 'playlist_mincount': 200,
4864 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4865 }, {
4866 'note': 'Playlist with unavailable videos in page 7',
4867 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4868 'info_dict': {
4869 'title': 'Uploads from BlankTV',
4870 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4871 'uploader': 'BlankTV',
4872 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4873 'channel': 'BlankTV',
4874 'channel_url': 'https://www.youtube.com/c/blanktv',
4875 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4876 'view_count': int,
4877 'tags': [],
4878 'uploader_url': 'https://www.youtube.com/c/blanktv',
4879 'modified_date': r're:\d{8}',
4880 'description': '',
4881 },
4882 'playlist_mincount': 1000,
4883 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4884 }, {
4885 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4886 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4887 'info_dict': {
4888 'title': 'Data Analysis with Dr Mike Pound',
4889 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4890 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4891 'uploader': 'Computerphile',
4892 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4893 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4894 'tags': [],
4895 'view_count': int,
4896 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4897 'channel_url': 'https://www.youtube.com/user/Computerphile',
4898 'channel': 'Computerphile',
4899 },
4900 'playlist_mincount': 11,
4901 }, {
4902 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4903 'only_matching': True,
4904 }, {
4905 'note': 'Playlist URL that does not actually serve a playlist',
4906 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4907 'info_dict': {
4908 'id': 'FqZTN594JQw',
4909 'ext': 'webm',
4910 'title': "Smiley's People 01 detective, Adventure Series, Action",
4911 'uploader': 'STREEM',
4912 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4913 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4914 'upload_date': '20150526',
4915 'license': 'Standard YouTube License',
4916 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4917 'categories': ['People & Blogs'],
4918 'tags': list,
4919 'view_count': int,
4920 'like_count': int,
4921 },
4922 'params': {
4923 'skip_download': True,
4924 },
4925 'skip': 'This video is not available.',
4926 'add_ie': [YoutubeIE.ie_key()],
4927 }, {
4928 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4929 'only_matching': True,
4930 }, {
4931 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4932 'only_matching': True,
4933 }, {
4934 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4935 'info_dict': {
4936 'id': 'GgL890LIznQ', # This will keep changing
4937 'ext': 'mp4',
4938 'title': str,
4939 'uploader': 'Sky News',
4940 'uploader_id': 'skynews',
4941 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4942 'upload_date': r're:\d{8}',
4943 'description': str,
4944 'categories': ['News & Politics'],
4945 'tags': list,
4946 'like_count': int,
4947 'release_timestamp': 1642502819,
4948 'channel': 'Sky News',
4949 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4950 'age_limit': 0,
4951 'view_count': int,
4952 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4953 'playable_in_embed': True,
4954 'release_date': '20220118',
4955 'availability': 'public',
4956 'live_status': 'is_live',
4957 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4958 'channel_follower_count': int
4959 },
4960 'params': {
4961 'skip_download': True,
4962 },
4963 'expected_warnings': ['Ignoring subtitle tracks found in '],
4964 }, {
4965 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4966 'info_dict': {
4967 'id': 'a48o2S1cPoo',
4968 'ext': 'mp4',
4969 'title': 'The Young Turks - Live Main Show',
4970 'uploader': 'The Young Turks',
4971 'uploader_id': 'TheYoungTurks',
4972 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4973 'upload_date': '20150715',
4974 'license': 'Standard YouTube License',
4975 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4976 'categories': ['News & Politics'],
4977 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4978 'like_count': int,
4979 },
4980 'params': {
4981 'skip_download': True,
4982 },
4983 'only_matching': True,
4984 }, {
4985 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
4986 'only_matching': True,
4987 }, {
4988 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
4989 'only_matching': True,
4990 }, {
4991 'note': 'A channel that is not live. Should raise error',
4992 'url': 'https://www.youtube.com/user/numberphile/live',
4993 'only_matching': True,
4994 }, {
4995 'url': 'https://www.youtube.com/feed/trending',
4996 'only_matching': True,
4997 }, {
4998 'url': 'https://www.youtube.com/feed/library',
4999 'only_matching': True,
5000 }, {
5001 'url': 'https://www.youtube.com/feed/history',
5002 'only_matching': True,
5003 }, {
5004 'url': 'https://www.youtube.com/feed/subscriptions',
5005 'only_matching': True,
5006 }, {
5007 'url': 'https://www.youtube.com/feed/watch_later',
5008 'only_matching': True,
5009 }, {
5010 'note': 'Recommended - redirects to home page.',
5011 'url': 'https://www.youtube.com/feed/recommended',
5012 'only_matching': True,
5013 }, {
5014 'note': 'inline playlist with not always working continuations',
5015 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5016 'only_matching': True,
5017 }, {
5018 'url': 'https://www.youtube.com/course',
5019 'only_matching': True,
5020 }, {
5021 'url': 'https://www.youtube.com/zsecurity',
5022 'only_matching': True,
5023 }, {
5024 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5025 'only_matching': True,
5026 }, {
5027 'url': 'https://www.youtube.com/TheYoungTurks/live',
5028 'only_matching': True,
5029 }, {
5030 'url': 'https://www.youtube.com/hashtag/cctv9',
5031 'info_dict': {
5032 'id': 'cctv9',
5033 'title': '#cctv9',
5034 'tags': [],
5035 },
5036 'playlist_mincount': 350,
5037 }, {
5038 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5039 'only_matching': True,
5040 }, {
5041 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5042 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5043 'only_matching': True
5044 }, {
5045 'note': '/browse/ should redirect to /channel/',
5046 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5047 'only_matching': True
5048 }, {
5049 'note': 'VLPL, should redirect to playlist?list=PL...',
5050 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5051 'info_dict': {
5052 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5053 'uploader': 'NoCopyrightSounds',
5054 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5055 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5056 'title': 'NCS Releases',
5057 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5058 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5059 'modified_date': r're:\d{8}',
5060 'view_count': int,
5061 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5062 'tags': [],
5063 'channel': 'NoCopyrightSounds',
5064 },
5065 'playlist_mincount': 166,
5066 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5067 }, {
5068 'note': 'Topic, should redirect to playlist?list=UU...',
5069 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5070 'info_dict': {
5071 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5072 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5073 'title': 'Uploads from Royalty Free Music - Topic',
5074 'uploader': 'Royalty Free Music - Topic',
5075 'tags': [],
5076 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5077 'channel': 'Royalty Free Music - Topic',
5078 'view_count': int,
5079 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5080 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5081 'modified_date': r're:\d{8}',
5082 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5083 'description': '',
5084 },
5085 'expected_warnings': [
5086 'The URL does not have a videos tab',
5087 r'[Uu]navailable videos (are|will be) hidden',
5088 ],
5089 'playlist_mincount': 101,
5090 }, {
5091 'note': 'Topic without a UU playlist',
5092 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5093 'info_dict': {
5094 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5095 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5096 'tags': [],
5097 },
5098 'expected_warnings': [
5099 'the playlist redirect gave error',
5100 ],
5101 'playlist_mincount': 9,
5102 }, {
5103 'note': 'Youtube music Album',
5104 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5105 'info_dict': {
5106 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5107 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5108 'tags': [],
5109 'view_count': int,
5110 'description': '',
5111 'availability': 'unlisted',
5112 'modified_date': r're:\d{8}',
5113 },
5114 'playlist_count': 50,
5115 }, {
5116 'note': 'unlisted single video playlist',
5117 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5118 'info_dict': {
5119 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5120 'uploader': 'colethedj',
5121 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5122 'title': 'yt-dlp unlisted playlist test',
5123 'availability': 'unlisted',
5124 'tags': [],
5125 'modified_date': '20211208',
5126 'channel': 'colethedj',
5127 'view_count': int,
5128 'description': '',
5129 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5130 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5131 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5132 },
5133 'playlist_count': 1,
5134 }, {
5135 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5136 'url': 'https://www.youtube.com/feed/recommended',
5137 'info_dict': {
5138 'id': 'recommended',
5139 'title': 'recommended',
5140 'tags': [],
5141 },
5142 'playlist_mincount': 50,
5143 'params': {
5144 'skip_download': True,
5145 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5146 },
5147 }, {
5148 'note': 'API Fallback: /videos tab, sorted by oldest first',
5149 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5150 'info_dict': {
5151 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5152 'title': 'Cody\'sLab - Videos',
5153 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5154 'uploader': 'Cody\'sLab',
5155 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5156 'channel': 'Cody\'sLab',
5157 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5158 'tags': [],
5159 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5160 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5161 'channel_follower_count': int
5162 },
5163 'playlist_mincount': 650,
5164 'params': {
5165 'skip_download': True,
5166 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5167 },
5168 }, {
5169 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5170 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5171 'info_dict': {
5172 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5173 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5174 'title': 'Uploads from Royalty Free Music - Topic',
5175 'uploader': 'Royalty Free Music - Topic',
5176 'modified_date': r're:\d{8}',
5177 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5178 'description': '',
5179 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5180 'tags': [],
5181 'channel': 'Royalty Free Music - Topic',
5182 'view_count': int,
5183 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5184 },
5185 'expected_warnings': [
5186 'does not have a videos tab',
5187 r'[Uu]navailable videos (are|will be) hidden',
5188 ],
5189 'playlist_mincount': 101,
5190 'params': {
5191 'skip_download': True,
5192 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5193 },
5194 }, {
5195 'note': 'non-standard redirect to regional channel',
5196 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5197 'only_matching': True
5198 }, {
5199 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5200 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5201 'info_dict': {
5202 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5203 'modified_date': '20220407',
5204 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5205 'tags': [],
5206 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5207 'uploader': 'pukkandan',
5208 'availability': 'unlisted',
5209 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5210 'channel': 'pukkandan',
5211 'description': 'Test for collaborative playlist',
5212 'title': 'yt-dlp test - collaborative playlist',
5213 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5214 },
5215 'playlist_mincount': 2
5216 }]
5217
5218 @classmethod
5219 def suitable(cls, url):
5220 return False if YoutubeIE.suitable(url) else super().suitable(url)
5221
5222 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5223
5224 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5225 def _real_extract(self, url, smuggled_data):
5226 item_id = self._match_id(url)
5227 url = compat_urlparse.urlunparse(
5228 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5229 compat_opts = self.get_param('compat_opts', [])
5230
5231 def get_mobj(url):
5232 mobj = self._URL_RE.match(url).groupdict()
5233 mobj.update((k, '') for k, v in mobj.items() if v is None)
5234 return mobj
5235
5236 mobj, redirect_warning = get_mobj(url), None
5237 # Youtube returns incomplete data if tabname is not lower case
5238 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5239 if is_channel:
5240 if smuggled_data.get('is_music_url'):
5241 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5242 item_id = item_id[2:]
5243 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5244 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5245 mdata = self._extract_tab_endpoint(
5246 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5247 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5248 get_all=False, expected_type=compat_str)
5249 if not murl:
5250 raise ExtractorError('Failed to resolve album to playlist')
5251 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5252 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5253 pre = f'https://www.youtube.com/channel/{item_id}'
5254
5255 original_tab_name = tab
5256 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5257 # Home URLs should redirect to /videos/
5258 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5259 'To download only the videos in the home page, add a "/featured" to the URL')
5260 tab = '/videos'
5261
5262 url = ''.join((pre, tab, post))
5263 mobj = get_mobj(url)
5264
5265 # Handle both video/playlist URLs
5266 qs = parse_qs(url)
5267 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5268
5269 if not video_id and mobj['not_channel'].startswith('watch'):
5270 if not playlist_id:
5271 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5272 raise ExtractorError('Unable to recognize tab page')
5273 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5274 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5275 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5276 mobj = get_mobj(url)
5277
5278 if video_id and playlist_id:
5279 if self.get_param('noplaylist'):
5280 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5281 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5282 ie=YoutubeIE.ie_key(), video_id=video_id)
5283 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5284
5285 data, ytcfg = self._extract_data(url, item_id)
5286
5287 # YouTube may provide a non-standard redirect to the regional channel
5288 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5289 redirect_url = traverse_obj(
5290 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5291 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5292 redirect_url = ''.join((
5293 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5294 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5295 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5296
5297 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5298 if tabs:
5299 selected_tab = self._extract_selected_tab(tabs)
5300 selected_tab_name = selected_tab.get('title', '').lower()
5301 if selected_tab_name == 'home':
5302 selected_tab_name = 'featured'
5303 requested_tab_name = mobj['tab'][1:]
5304 if 'no-youtube-channel-redirect' not in compat_opts:
5305 if requested_tab_name == 'live':
5306 # Live tab should have redirected to the video
5307 raise ExtractorError('The channel is not currently live', expected=True)
5308 if requested_tab_name not in ('', selected_tab_name):
5309 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5310 if not original_tab_name:
5311 if item_id[:2] == 'UC':
5312 # Topic channels don't have /videos. Use the equivalent playlist instead
5313 pl_id = f'UU{item_id[2:]}'
5314 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5315 try:
5316 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5317 except ExtractorError:
5318 redirect_warning += ' and the playlist redirect gave error'
5319 else:
5320 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5321 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5322 if selected_tab_name and selected_tab_name != requested_tab_name:
5323 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5324 else:
5325 raise ExtractorError(redirect_warning, expected=True)
5326
5327 if redirect_warning:
5328 self.to_screen(redirect_warning)
5329 self.write_debug(f'Final URL: {url}')
5330
5331 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5332 if 'no-youtube-unavailable-videos' not in compat_opts:
5333 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5334 self._extract_and_report_alerts(data, only_once=True)
5335 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5336 if tabs:
5337 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5338
5339 playlist = traverse_obj(
5340 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5341 if playlist:
5342 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5343
5344 video_id = traverse_obj(
5345 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5346 if video_id:
5347 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5348 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5349 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5350 ie=YoutubeIE.ie_key(), video_id=video_id)
5351
5352 raise ExtractorError('Unable to recognize tab page')
5353
5354
5355class YoutubePlaylistIE(InfoExtractor):
5356 IE_DESC = 'YouTube playlists'
5357 _VALID_URL = r'''(?x)(?:
5358 (?:https?://)?
5359 (?:\w+\.)?
5360 (?:
5361 (?:
5362 youtube(?:kids)?\.com|
5363 %(invidious)s
5364 )
5365 /.*?\?.*?\blist=
5366 )?
5367 (?P<id>%(playlist_id)s)
5368 )''' % {
5369 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5370 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5371 }
5372 IE_NAME = 'youtube:playlist'
5373 _TESTS = [{
5374 'note': 'issue #673',
5375 'url': 'PLBB231211A4F62143',
5376 'info_dict': {
5377 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5378 'id': 'PLBB231211A4F62143',
5379 'uploader': 'Wickman',
5380 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5381 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5382 'view_count': int,
5383 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5384 'modified_date': r're:\d{8}',
5385 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5386 'channel': 'Wickman',
5387 'tags': [],
5388 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5389 },
5390 'playlist_mincount': 29,
5391 }, {
5392 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5393 'info_dict': {
5394 'title': 'YDL_safe_search',
5395 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5396 },
5397 'playlist_count': 2,
5398 'skip': 'This playlist is private',
5399 }, {
5400 'note': 'embedded',
5401 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5402 'playlist_count': 4,
5403 'info_dict': {
5404 'title': 'JODA15',
5405 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5406 'uploader': 'milan',
5407 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5408 'description': '',
5409 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5410 'tags': [],
5411 'modified_date': '20140919',
5412 'view_count': int,
5413 'channel': 'milan',
5414 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5415 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5416 },
5417 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5418 }, {
5419 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5420 'playlist_mincount': 654,
5421 'info_dict': {
5422 'title': '2018 Chinese New Singles (11/6 updated)',
5423 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5424 'uploader': 'LBK',
5425 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5426 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5427 'channel': 'LBK',
5428 'view_count': int,
5429 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5430 'tags': [],
5431 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5432 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5433 'modified_date': r're:\d{8}',
5434 },
5435 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5436 }, {
5437 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5438 'only_matching': True,
5439 }, {
5440 # music album playlist
5441 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5442 'only_matching': True,
5443 }]
5444
5445 @classmethod
5446 def suitable(cls, url):
5447 if YoutubeTabIE.suitable(url):
5448 return False
5449 from ..utils import parse_qs
5450 qs = parse_qs(url)
5451 if qs.get('v', [None])[0]:
5452 return False
5453 return super().suitable(url)
5454
5455 def _real_extract(self, url):
5456 playlist_id = self._match_id(url)
5457 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5458 url = update_url_query(
5459 'https://www.youtube.com/playlist',
5460 parse_qs(url) or {'list': playlist_id})
5461 if is_music_url:
5462 url = smuggle_url(url, {'is_music_url': True})
5463 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5464
5465
5466class YoutubeYtBeIE(InfoExtractor):
5467 IE_DESC = 'youtu.be'
5468 _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}
5469 _TESTS = [{
5470 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5471 'info_dict': {
5472 'id': 'yeWKywCrFtk',
5473 'ext': 'mp4',
5474 'title': 'Small Scale Baler and Braiding Rugs',
5475 'uploader': 'Backus-Page House Museum',
5476 'uploader_id': 'backuspagemuseum',
5477 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5478 'upload_date': '20161008',
5479 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5480 'categories': ['Nonprofits & Activism'],
5481 'tags': list,
5482 'like_count': int,
5483 'age_limit': 0,
5484 'playable_in_embed': True,
5485 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5486 'channel': 'Backus-Page House Museum',
5487 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5488 'live_status': 'not_live',
5489 'view_count': int,
5490 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5491 'availability': 'public',
5492 'duration': 59,
5493 },
5494 'params': {
5495 'noplaylist': True,
5496 'skip_download': True,
5497 },
5498 }, {
5499 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5500 'only_matching': True,
5501 }]
5502
5503 def _real_extract(self, url):
5504 mobj = self._match_valid_url(url)
5505 video_id = mobj.group('id')
5506 playlist_id = mobj.group('playlist_id')
5507 return self.url_result(
5508 update_url_query('https://www.youtube.com/watch', {
5509 'v': video_id,
5510 'list': playlist_id,
5511 'feature': 'youtu.be',
5512 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5513
5514
5515class YoutubeLivestreamEmbedIE(InfoExtractor):
5516 IE_DESC = 'YouTube livestream embeds'
5517 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5518 _TESTS = [{
5519 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5520 'only_matching': True,
5521 }]
5522
5523 def _real_extract(self, url):
5524 channel_id = self._match_id(url)
5525 return self.url_result(
5526 f'https://www.youtube.com/channel/{channel_id}/live',
5527 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5528
5529
5530class YoutubeYtUserIE(InfoExtractor):
5531 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5532 IE_NAME = 'youtube:user'
5533 _VALID_URL = r'ytuser:(?P<id>.+)'
5534 _TESTS = [{
5535 'url': 'ytuser:phihag',
5536 'only_matching': True,
5537 }]
5538
5539 def _real_extract(self, url):
5540 user_id = self._match_id(url)
5541 return self.url_result(
5542 'https://www.youtube.com/user/%s/videos' % user_id,
5543 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5544
5545
5546class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5547 IE_NAME = 'youtube:favorites'
5548 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5549 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5550 _LOGIN_REQUIRED = True
5551 _TESTS = [{
5552 'url': ':ytfav',
5553 'only_matching': True,
5554 }, {
5555 'url': ':ytfavorites',
5556 'only_matching': True,
5557 }]
5558
5559 def _real_extract(self, url):
5560 return self.url_result(
5561 'https://www.youtube.com/playlist?list=LL',
5562 ie=YoutubeTabIE.ie_key())
5563
5564
5565class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5566 IE_NAME = 'youtube:notif'
5567 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5568 _VALID_URL = r':ytnotif(?:ication)?s?'
5569 _LOGIN_REQUIRED = True
5570 _TESTS = [{
5571 'url': ':ytnotif',
5572 'only_matching': True,
5573 }, {
5574 'url': ':ytnotifications',
5575 'only_matching': True,
5576 }]
5577
5578 def _extract_notification_menu(self, response, continuation_list):
5579 notification_list = traverse_obj(
5580 response,
5581 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5582 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5583 expected_type=list) or []
5584 continuation_list[0] = None
5585 for item in notification_list:
5586 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5587 if entry:
5588 yield entry
5589 continuation = item.get('continuationItemRenderer')
5590 if continuation:
5591 continuation_list[0] = continuation
5592
5593 def _extract_notification_renderer(self, notification):
5594 video_id = traverse_obj(
5595 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5596 url = f'https://www.youtube.com/watch?v={video_id}'
5597 channel_id = None
5598 if not video_id:
5599 browse_ep = traverse_obj(
5600 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5601 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5602 post_id = self._search_regex(
5603 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5604 'post id', default=None)
5605 if not channel_id or not post_id:
5606 return
5607 # The direct /post url redirects to this in the browser
5608 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5609
5610 channel = traverse_obj(
5611 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5612 expected_type=str)
5613 title = self._search_regex(
5614 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5615 'video title', default=None)
5616 if title:
5617 title = title.replace('\xad', '') # remove soft hyphens
5618 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5619 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5620 else None)
5621 return {
5622 '_type': 'url',
5623 'url': url,
5624 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5625 'video_id': video_id,
5626 'title': title,
5627 'channel_id': channel_id,
5628 'channel': channel,
5629 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5630 'upload_date': upload_date,
5631 }
5632
5633 def _notification_menu_entries(self, ytcfg):
5634 continuation_list = [None]
5635 response = None
5636 for page in itertools.count(1):
5637 ctoken = traverse_obj(
5638 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5639 response = self._extract_response(
5640 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5641 ep='notification/get_notification_menu', check_get_keys='actions',
5642 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5643 yield from self._extract_notification_menu(response, continuation_list)
5644 if not continuation_list[0]:
5645 break
5646
5647 def _real_extract(self, url):
5648 display_id = 'notifications'
5649 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5650 self._report_playlist_authcheck(ytcfg)
5651 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5652
5653
5654class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5655 IE_DESC = 'YouTube search'
5656 IE_NAME = 'youtube:search'
5657 _SEARCH_KEY = 'ytsearch'
5658 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5659 _TESTS = [{
5660 'url': 'ytsearch5:youtube-dl test video',
5661 'playlist_count': 5,
5662 'info_dict': {
5663 'id': 'youtube-dl test video',
5664 'title': 'youtube-dl test video',
5665 }
5666 }]
5667
5668
5669class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5670 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5671 _SEARCH_KEY = 'ytsearchdate'
5672 IE_DESC = 'YouTube search, newest videos first'
5673 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5674 _TESTS = [{
5675 'url': 'ytsearchdate5:youtube-dl test video',
5676 'playlist_count': 5,
5677 'info_dict': {
5678 'id': 'youtube-dl test video',
5679 'title': 'youtube-dl test video',
5680 }
5681 }]
5682
5683
5684class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5685 IE_DESC = 'YouTube search URLs with sorting and filter support'
5686 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5687 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5688 _TESTS = [{
5689 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5690 'playlist_mincount': 5,
5691 'info_dict': {
5692 'id': 'youtube-dl test video',
5693 'title': 'youtube-dl test video',
5694 }
5695 }, {
5696 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5697 'playlist_mincount': 5,
5698 'info_dict': {
5699 'id': 'python',
5700 'title': 'python',
5701 }
5702 }, {
5703 'url': 'https://www.youtube.com/results?search_query=%23cats',
5704 'playlist_mincount': 1,
5705 'info_dict': {
5706 'id': '#cats',
5707 'title': '#cats',
5708 'entries': [{
5709 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5710 'title': '#cats',
5711 }],
5712 },
5713 }, {
5714 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5715 'only_matching': True,
5716 }]
5717
5718 def _real_extract(self, url):
5719 qs = parse_qs(url)
5720 query = (qs.get('search_query') or qs.get('q'))[0]
5721 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5722
5723
5724class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5725 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5726 IE_NAME = 'youtube:music:search_url'
5727 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5728 _TESTS = [{
5729 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5730 'playlist_count': 16,
5731 'info_dict': {
5732 'id': 'royalty free music',
5733 'title': 'royalty free music',
5734 }
5735 }, {
5736 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5737 'playlist_mincount': 30,
5738 'info_dict': {
5739 'id': 'royalty free music - songs',
5740 'title': 'royalty free music - songs',
5741 },
5742 'params': {'extract_flat': 'in_playlist'}
5743 }, {
5744 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5745 'playlist_mincount': 30,
5746 'info_dict': {
5747 'id': 'royalty free music - community playlists',
5748 'title': 'royalty free music - community playlists',
5749 },
5750 'params': {'extract_flat': 'in_playlist'}
5751 }]
5752
5753 _SECTIONS = {
5754 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5755 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5756 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5757 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5758 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5759 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5760 }
5761
5762 def _real_extract(self, url):
5763 qs = parse_qs(url)
5764 query = (qs.get('search_query') or qs.get('q'))[0]
5765 params = qs.get('sp', (None,))[0]
5766 if params:
5767 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5768 else:
5769 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5770 params = self._SECTIONS.get(section)
5771 if not params:
5772 section = None
5773 title = join_nonempty(query, section, delim=' - ')
5774 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5775
5776
5777class YoutubeFeedsInfoExtractor(InfoExtractor):
5778 """
5779 Base class for feed extractors
5780 Subclasses must define the _FEED_NAME property.
5781 """
5782 _LOGIN_REQUIRED = True
5783
5784 def _real_initialize(self):
5785 YoutubeBaseInfoExtractor._check_login_required(self)
5786
5787 @property
5788 def IE_NAME(self):
5789 return 'youtube:%s' % self._FEED_NAME
5790
5791 def _real_extract(self, url):
5792 return self.url_result(
5793 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5794
5795
5796class YoutubeWatchLaterIE(InfoExtractor):
5797 IE_NAME = 'youtube:watchlater'
5798 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5799 _VALID_URL = r':ytwatchlater'
5800 _TESTS = [{
5801 'url': ':ytwatchlater',
5802 'only_matching': True,
5803 }]
5804
5805 def _real_extract(self, url):
5806 return self.url_result(
5807 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5808
5809
5810class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5811 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5812 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5813 _FEED_NAME = 'recommended'
5814 _LOGIN_REQUIRED = False
5815 _TESTS = [{
5816 'url': ':ytrec',
5817 'only_matching': True,
5818 }, {
5819 'url': ':ytrecommended',
5820 'only_matching': True,
5821 }, {
5822 'url': 'https://youtube.com',
5823 'only_matching': True,
5824 }]
5825
5826
5827class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5828 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5829 _VALID_URL = r':ytsub(?:scription)?s?'
5830 _FEED_NAME = 'subscriptions'
5831 _TESTS = [{
5832 'url': ':ytsubs',
5833 'only_matching': True,
5834 }, {
5835 'url': ':ytsubscriptions',
5836 'only_matching': True,
5837 }]
5838
5839
5840class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5841 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5842 _VALID_URL = r':ythis(?:tory)?'
5843 _FEED_NAME = 'history'
5844 _TESTS = [{
5845 'url': ':ythistory',
5846 'only_matching': True,
5847 }]
5848
5849
5850class YoutubeStoriesIE(InfoExtractor):
5851 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5852 IE_NAME = 'youtube:stories'
5853 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5854 _TESTS = [{
5855 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5856 'only_matching': True,
5857 }]
5858
5859 def _real_extract(self, url):
5860 playlist_id = f'RLTD{self._match_id(url)}'
5861 return self.url_result(
5862 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5863 ie=YoutubeTabIE, video_id=playlist_id)
5864
5865
5866class YoutubeTruncatedURLIE(InfoExtractor):
5867 IE_NAME = 'youtube:truncated_url'
5868 IE_DESC = False # Do not list
5869 _VALID_URL = r'''(?x)
5870 (?:https?://)?
5871 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5872 (?:watch\?(?:
5873 feature=[a-z_]+|
5874 annotation_id=annotation_[^&]+|
5875 x-yt-cl=[0-9]+|
5876 hl=[^&]*|
5877 t=[0-9]+
5878 )?
5879 |
5880 attribution_link\?a=[^&]+
5881 )
5882 $
5883 '''
5884
5885 _TESTS = [{
5886 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5887 'only_matching': True,
5888 }, {
5889 'url': 'https://www.youtube.com/watch?',
5890 'only_matching': True,
5891 }, {
5892 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5893 'only_matching': True,
5894 }, {
5895 'url': 'https://www.youtube.com/watch?feature=foo',
5896 'only_matching': True,
5897 }, {
5898 'url': 'https://www.youtube.com/watch?hl=en-GB',
5899 'only_matching': True,
5900 }, {
5901 'url': 'https://www.youtube.com/watch?t=2372',
5902 'only_matching': True,
5903 }]
5904
5905 def _real_extract(self, url):
5906 raise ExtractorError(
5907 'Did you forget to quote the URL? Remember that & is a meta '
5908 'character in most shells, so you want to put the URL in quotes, '
5909 'like youtube-dl '
5910 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5911 ' or simply youtube-dl BaW_jenozKc .',
5912 expected=True)
5913
5914
5915class YoutubeClipIE(InfoExtractor):
5916 IE_NAME = 'youtube:clip'
5917 IE_DESC = False # Do not list
5918 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5919
5920 def _real_extract(self, url):
5921 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5922 return self.url_result(url, 'Generic')
5923
5924
5925class YoutubeTruncatedIDIE(InfoExtractor):
5926 IE_NAME = 'youtube:truncated_id'
5927 IE_DESC = False # Do not list
5928 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5929
5930 _TESTS = [{
5931 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5932 'only_matching': True,
5933 }]
5934
5935 def _real_extract(self, url):
5936 video_id = self._match_id(url)
5937 raise ExtractorError(
5938 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5939 expected=True)