]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[youtube] Differentiate descriptive audio by language code
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import calendar
6 import copy
7 import datetime
8 import functools
9 import hashlib
10 import itertools
11 import json
12 import math
13 import os.path
14 import random
15 import re
16 import sys
17 import time
18 import traceback
19 import threading
20
21 from .common import InfoExtractor, SearchInfoExtractor
22 from ..compat import (
23 compat_chr,
24 compat_HTTPError,
25 compat_parse_qs,
26 compat_str,
27 compat_urllib_parse_unquote_plus,
28 compat_urllib_parse_urlencode,
29 compat_urllib_parse_urlparse,
30 compat_urlparse,
31 )
32 from ..jsinterp import JSInterpreter
33 from ..utils import (
34 bug_reports_message,
35 clean_html,
36 datetime_from_str,
37 dict_get,
38 error_to_compat_str,
39 ExtractorError,
40 float_or_none,
41 format_field,
42 int_or_none,
43 is_html,
44 join_nonempty,
45 js_to_json,
46 mimetype2ext,
47 network_exceptions,
48 NO_DEFAULT,
49 orderedSet,
50 parse_codecs,
51 parse_count,
52 parse_duration,
53 parse_iso8601,
54 parse_qs,
55 qualities,
56 remove_end,
57 remove_start,
58 smuggle_url,
59 str_or_none,
60 str_to_int,
61 strftime_or_none,
62 traverse_obj,
63 try_get,
64 unescapeHTML,
65 unified_strdate,
66 unified_timestamp,
67 unsmuggle_url,
68 update_url_query,
69 url_or_none,
70 urljoin,
71 variadic,
72 )
73
74
75 def get_first(obj, keys, **kwargs):
76 return traverse_obj(obj, (..., *variadic(keys)), **kwargs, get_all=False)
77
78
79 # any clients starting with _ cannot be explicity requested by the user
80 INNERTUBE_CLIENTS = {
81 'web': {
82 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
83 'INNERTUBE_CONTEXT': {
84 'client': {
85 'clientName': 'WEB',
86 'clientVersion': '2.20211221.00.00',
87 }
88 },
89 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
90 },
91 'web_embedded': {
92 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
93 'INNERTUBE_CONTEXT': {
94 'client': {
95 'clientName': 'WEB_EMBEDDED_PLAYER',
96 'clientVersion': '1.20211215.00.01',
97 },
98 },
99 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
100 },
101 'web_music': {
102 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
103 'INNERTUBE_HOST': 'music.youtube.com',
104 'INNERTUBE_CONTEXT': {
105 'client': {
106 'clientName': 'WEB_REMIX',
107 'clientVersion': '1.20211213.00.00',
108 }
109 },
110 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
111 },
112 'web_creator': {
113 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
114 'INNERTUBE_CONTEXT': {
115 'client': {
116 'clientName': 'WEB_CREATOR',
117 'clientVersion': '1.20211220.02.00',
118 }
119 },
120 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
121 },
122 'android': {
123 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
124 'INNERTUBE_CONTEXT': {
125 'client': {
126 'clientName': 'ANDROID',
127 'clientVersion': '16.49',
128 }
129 },
130 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
131 'REQUIRE_JS_PLAYER': False
132 },
133 'android_embedded': {
134 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
135 'INNERTUBE_CONTEXT': {
136 'client': {
137 'clientName': 'ANDROID_EMBEDDED_PLAYER',
138 'clientVersion': '16.49',
139 },
140 },
141 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
142 'REQUIRE_JS_PLAYER': False
143 },
144 'android_music': {
145 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
146 'INNERTUBE_CONTEXT': {
147 'client': {
148 'clientName': 'ANDROID_MUSIC',
149 'clientVersion': '4.57',
150 }
151 },
152 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
153 'REQUIRE_JS_PLAYER': False
154 },
155 'android_creator': {
156 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
157 'INNERTUBE_CONTEXT': {
158 'client': {
159 'clientName': 'ANDROID_CREATOR',
160 'clientVersion': '21.47',
161 },
162 },
163 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
164 'REQUIRE_JS_PLAYER': False
165 },
166 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
167 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
168 'ios': {
169 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
170 'INNERTUBE_CONTEXT': {
171 'client': {
172 'clientName': 'IOS',
173 'clientVersion': '16.46',
174 'deviceModel': 'iPhone14,3',
175 }
176 },
177 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
178 'REQUIRE_JS_PLAYER': False
179 },
180 'ios_embedded': {
181 'INNERTUBE_CONTEXT': {
182 'client': {
183 'clientName': 'IOS_MESSAGES_EXTENSION',
184 'clientVersion': '16.46',
185 'deviceModel': 'iPhone14,3',
186 },
187 },
188 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
189 'REQUIRE_JS_PLAYER': False
190 },
191 'ios_music': {
192 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
193 'INNERTUBE_CONTEXT': {
194 'client': {
195 'clientName': 'IOS_MUSIC',
196 'clientVersion': '4.57',
197 },
198 },
199 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
200 'REQUIRE_JS_PLAYER': False
201 },
202 'ios_creator': {
203 'INNERTUBE_CONTEXT': {
204 'client': {
205 'clientName': 'IOS_CREATOR',
206 'clientVersion': '21.47',
207 },
208 },
209 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
210 'REQUIRE_JS_PLAYER': False
211 },
212 # mweb has 'ultralow' formats
213 # See: https://github.com/yt-dlp/yt-dlp/pull/557
214 'mweb': {
215 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
216 'INNERTUBE_CONTEXT': {
217 'client': {
218 'clientName': 'MWEB',
219 'clientVersion': '2.20211221.01.00',
220 }
221 },
222 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
223 }
224 }
225
226
227 def build_innertube_clients():
228 third_party = {
229 'embedUrl': 'https://google.com', # Can be any valid URL
230 }
231 base_clients = ('android', 'web', 'ios', 'mweb')
232 priority = qualities(base_clients[::-1])
233
234 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
235 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
236 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
237 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
238 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
239 ytcfg['priority'] = 10 * priority(client.split('_', 1)[0])
240
241 if client in base_clients:
242 INNERTUBE_CLIENTS[f'{client}_agegate'] = agegate_ytcfg = copy.deepcopy(ytcfg)
243 agegate_ytcfg['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
244 agegate_ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party
245 agegate_ytcfg['priority'] -= 1
246 elif client.endswith('_embedded'):
247 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party
248 ytcfg['priority'] -= 2
249 else:
250 ytcfg['priority'] -= 3
251
252
253 build_innertube_clients()
254
255
256 class YoutubeBaseInfoExtractor(InfoExtractor):
257 """Provide base functions for Youtube extractors"""
258
259 _RESERVED_NAMES = (
260 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
261 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
262 r'browse|oembed|get_video_info|iframe_api|s/player|'
263 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
264
265 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
266
267 _NETRC_MACHINE = 'youtube'
268
269 # If True it will raise an error if no login info is provided
270 _LOGIN_REQUIRED = False
271
272 _INVIDIOUS_SITES = (
273 # invidious-redirect websites
274 r'(?:www\.)?redirect\.invidious\.io',
275 r'(?:(?:www|dev)\.)?invidio\.us',
276 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
277 r'(?:www\.)?invidious\.pussthecat\.org',
278 r'(?:www\.)?invidious\.zee\.li',
279 r'(?:www\.)?invidious\.ethibox\.fr',
280 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
281 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
282 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
283 # youtube-dl invidious instances list
284 r'(?:(?:www|no)\.)?invidiou\.sh',
285 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
286 r'(?:www\.)?invidious\.kabi\.tk',
287 r'(?:www\.)?invidious\.mastodon\.host',
288 r'(?:www\.)?invidious\.zapashcanon\.fr',
289 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
290 r'(?:www\.)?invidious\.tinfoil-hat\.net',
291 r'(?:www\.)?invidious\.himiko\.cloud',
292 r'(?:www\.)?invidious\.reallyancient\.tech',
293 r'(?:www\.)?invidious\.tube',
294 r'(?:www\.)?invidiou\.site',
295 r'(?:www\.)?invidious\.site',
296 r'(?:www\.)?invidious\.xyz',
297 r'(?:www\.)?invidious\.nixnet\.xyz',
298 r'(?:www\.)?invidious\.048596\.xyz',
299 r'(?:www\.)?invidious\.drycat\.fr',
300 r'(?:www\.)?inv\.skyn3t\.in',
301 r'(?:www\.)?tube\.poal\.co',
302 r'(?:www\.)?tube\.connect\.cafe',
303 r'(?:www\.)?vid\.wxzm\.sx',
304 r'(?:www\.)?vid\.mint\.lgbt',
305 r'(?:www\.)?vid\.puffyan\.us',
306 r'(?:www\.)?yewtu\.be',
307 r'(?:www\.)?yt\.elukerio\.org',
308 r'(?:www\.)?yt\.lelux\.fi',
309 r'(?:www\.)?invidious\.ggc-project\.de',
310 r'(?:www\.)?yt\.maisputain\.ovh',
311 r'(?:www\.)?ytprivate\.com',
312 r'(?:www\.)?invidious\.13ad\.de',
313 r'(?:www\.)?invidious\.toot\.koeln',
314 r'(?:www\.)?invidious\.fdn\.fr',
315 r'(?:www\.)?watch\.nettohikari\.com',
316 r'(?:www\.)?invidious\.namazso\.eu',
317 r'(?:www\.)?invidious\.silkky\.cloud',
318 r'(?:www\.)?invidious\.exonip\.de',
319 r'(?:www\.)?invidious\.riverside\.rocks',
320 r'(?:www\.)?invidious\.blamefran\.net',
321 r'(?:www\.)?invidious\.moomoo\.de',
322 r'(?:www\.)?ytb\.trom\.tf',
323 r'(?:www\.)?yt\.cyberhost\.uk',
324 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
325 r'(?:www\.)?qklhadlycap4cnod\.onion',
326 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
327 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
328 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
329 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
330 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
331 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
332 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
333 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
334 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
335 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
336 )
337
338 def _login(self):
339 """
340 Attempt to log in to YouTube.
341 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
342 """
343
344 if (self._LOGIN_REQUIRED
345 and self.get_param('cookiefile') is None
346 and self.get_param('cookiesfrombrowser') is None):
347 self.raise_login_required(
348 'Login details are needed to download this content', method='cookies')
349 username, password = self._get_login_info()
350 if username:
351 self.report_warning(f'Cannot login to YouTube using username and password. {self._LOGIN_HINTS["cookies"]}')
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._login()
384
385 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
386 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
387 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
388
389 def _get_default_ytcfg(self, client='web'):
390 return copy.deepcopy(INNERTUBE_CLIENTS[client])
391
392 def _get_innertube_host(self, client='web'):
393 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
394
395 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
396 # try_get but with fallback to default ytcfg client values when present
397 _func = lambda y: try_get(y, getter, expected_type)
398 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
399
400 def _extract_client_name(self, ytcfg, default_client='web'):
401 return self._ytcfg_get_safe(
402 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
403 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
404
405 def _extract_client_version(self, ytcfg, default_client='web'):
406 return self._ytcfg_get_safe(
407 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
408 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
409
410 def _extract_api_key(self, ytcfg=None, default_client='web'):
411 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
412
413 def _extract_context(self, ytcfg=None, default_client='web'):
414 context = get_first(
415 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
416 # Enforce language and tz for extraction
417 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
418 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
419 return context
420
421 _SAPISID = None
422
423 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
424 time_now = round(time.time())
425 if self._SAPISID is None:
426 yt_cookies = self._get_cookies('https://www.youtube.com')
427 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
428 # See: https://github.com/yt-dlp/yt-dlp/issues/393
429 sapisid_cookie = dict_get(
430 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
431 if sapisid_cookie and sapisid_cookie.value:
432 self._SAPISID = sapisid_cookie.value
433 self.write_debug('Extracted SAPISID cookie')
434 # SAPISID cookie is required if not already present
435 if not yt_cookies.get('SAPISID'):
436 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
437 self._set_cookie(
438 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
439 else:
440 self._SAPISID = False
441 if not self._SAPISID:
442 return None
443 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
444 sapisidhash = hashlib.sha1(
445 f'{time_now} {self._SAPISID} {origin}'.encode('utf-8')).hexdigest()
446 return f'SAPISIDHASH {time_now}_{sapisidhash}'
447
448 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
449 note='Downloading API JSON', errnote='Unable to download API page',
450 context=None, api_key=None, api_hostname=None, default_client='web'):
451
452 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
453 data.update(query)
454 real_headers = self.generate_api_headers(default_client=default_client)
455 real_headers.update({'content-type': 'application/json'})
456 if headers:
457 real_headers.update(headers)
458 return self._download_json(
459 'https://%s/youtubei/v1/%s' % (api_hostname or self._get_innertube_host(default_client), ep),
460 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
461 data=json.dumps(data).encode('utf8'), headers=real_headers,
462 query={'key': api_key or self._extract_api_key()})
463
464 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
465 data = self._search_regex(
466 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
467 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
468 if data:
469 return self._parse_json(data, item_id, fatal=fatal)
470
471 @staticmethod
472 def _extract_session_index(*data):
473 """
474 Index of current account in account list.
475 See: https://github.com/yt-dlp/yt-dlp/pull/519
476 """
477 for ytcfg in data:
478 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
479 if session_index is not None:
480 return session_index
481
482 # Deprecated?
483 def _extract_identity_token(self, ytcfg=None, webpage=None):
484 if ytcfg:
485 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
486 if token:
487 return token
488 if webpage:
489 return self._search_regex(
490 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
491 'identity token', default=None, fatal=False)
492
493 @staticmethod
494 def _extract_account_syncid(*args):
495 """
496 Extract syncId required to download private playlists of secondary channels
497 @params response and/or ytcfg
498 """
499 for data in args:
500 # ytcfg includes channel_syncid if on secondary channel
501 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
502 if delegated_sid:
503 return delegated_sid
504 sync_ids = (try_get(
505 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
506 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
507 if len(sync_ids) >= 2 and sync_ids[1]:
508 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
509 # and just "user_syncid||" for primary channel. We only want the channel_syncid
510 return sync_ids[0]
511
512 @staticmethod
513 def _extract_visitor_data(*args):
514 """
515 Extracts visitorData from an API response or ytcfg
516 Appears to be used to track session state
517 """
518 return get_first(
519 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
520 expected_type=str)
521
522 @property
523 def is_authenticated(self):
524 return bool(self._generate_sapisidhash_header())
525
526 def extract_ytcfg(self, video_id, webpage):
527 if not webpage:
528 return {}
529 return self._parse_json(
530 self._search_regex(
531 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
532 default='{}'), video_id, fatal=False) or {}
533
534 def generate_api_headers(
535 self, *, ytcfg=None, account_syncid=None, session_index=None,
536 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
537
538 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
539 headers = {
540 'X-YouTube-Client-Name': compat_str(
541 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
542 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
543 'Origin': origin,
544 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
545 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
546 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
547 }
548 if session_index is None:
549 session_index = self._extract_session_index(ytcfg)
550 if account_syncid or session_index is not None:
551 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
552
553 auth = self._generate_sapisidhash_header(origin)
554 if auth is not None:
555 headers['Authorization'] = auth
556 headers['X-Origin'] = origin
557 return {h: v for h, v in headers.items() if v is not None}
558
559 @staticmethod
560 def _build_api_continuation_query(continuation, ctp=None):
561 query = {
562 'continuation': continuation
563 }
564 # TODO: Inconsistency with clickTrackingParams.
565 # Currently we have a fixed ctp contained within context (from ytcfg)
566 # and a ctp in root query for continuation.
567 if ctp:
568 query['clickTracking'] = {'clickTrackingParams': ctp}
569 return query
570
571 @classmethod
572 def _extract_next_continuation_data(cls, renderer):
573 next_continuation = try_get(
574 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
575 lambda x: x['continuation']['reloadContinuationData']), dict)
576 if not next_continuation:
577 return
578 continuation = next_continuation.get('continuation')
579 if not continuation:
580 return
581 ctp = next_continuation.get('clickTrackingParams')
582 return cls._build_api_continuation_query(continuation, ctp)
583
584 @classmethod
585 def _extract_continuation_ep_data(cls, continuation_ep: dict):
586 if isinstance(continuation_ep, dict):
587 continuation = try_get(
588 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
589 if not continuation:
590 return
591 ctp = continuation_ep.get('clickTrackingParams')
592 return cls._build_api_continuation_query(continuation, ctp)
593
594 @classmethod
595 def _extract_continuation(cls, renderer):
596 next_continuation = cls._extract_next_continuation_data(renderer)
597 if next_continuation:
598 return next_continuation
599
600 contents = []
601 for key in ('contents', 'items'):
602 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
603
604 for content in contents:
605 if not isinstance(content, dict):
606 continue
607 continuation_ep = try_get(
608 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
609 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
610 dict)
611 continuation = cls._extract_continuation_ep_data(continuation_ep)
612 if continuation:
613 return continuation
614
615 @classmethod
616 def _extract_alerts(cls, data):
617 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
618 if not isinstance(alert_dict, dict):
619 continue
620 for alert in alert_dict.values():
621 alert_type = alert.get('type')
622 if not alert_type:
623 continue
624 message = cls._get_text(alert, 'text')
625 if message:
626 yield alert_type, message
627
628 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
629 errors = []
630 warnings = []
631 for alert_type, alert_message in alerts:
632 if alert_type.lower() == 'error' and fatal:
633 errors.append([alert_type, alert_message])
634 else:
635 warnings.append([alert_type, alert_message])
636
637 for alert_type, alert_message in (warnings + errors[:-1]):
638 self.report_warning('YouTube said: %s - %s' % (alert_type, alert_message), only_once=only_once)
639 if errors:
640 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
641
642 def _extract_and_report_alerts(self, data, *args, **kwargs):
643 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
644
645 def _extract_badges(self, renderer: dict):
646 badges = set()
647 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
648 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
649 if label:
650 badges.add(label.lower())
651 return badges
652
653 @staticmethod
654 def _get_text(data, *path_list, max_runs=None):
655 for path in path_list or [None]:
656 if path is None:
657 obj = [data]
658 else:
659 obj = traverse_obj(data, path, default=[])
660 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
661 obj = [obj]
662 for item in obj:
663 text = try_get(item, lambda x: x['simpleText'], compat_str)
664 if text:
665 return text
666 runs = try_get(item, lambda x: x['runs'], list) or []
667 if not runs and isinstance(item, list):
668 runs = item
669
670 runs = runs[:min(len(runs), max_runs or len(runs))]
671 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
672 if text:
673 return text
674
675 def _get_count(self, data, *path_list):
676 count_text = self._get_text(data, *path_list) or ''
677 count = parse_count(count_text)
678 if count is None:
679 count = str_to_int(
680 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
681 return count
682
683 @staticmethod
684 def _extract_thumbnails(data, *path_list):
685 """
686 Extract thumbnails from thumbnails dict
687 @param path_list: path list to level that contains 'thumbnails' key
688 """
689 thumbnails = []
690 for path in path_list or [()]:
691 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
692 thumbnail_url = url_or_none(thumbnail.get('url'))
693 if not thumbnail_url:
694 continue
695 # Sometimes youtube gives a wrong thumbnail URL. See:
696 # https://github.com/yt-dlp/yt-dlp/issues/233
697 # https://github.com/ytdl-org/youtube-dl/issues/28023
698 if 'maxresdefault' in thumbnail_url:
699 thumbnail_url = thumbnail_url.split('?')[0]
700 thumbnails.append({
701 'url': thumbnail_url,
702 'height': int_or_none(thumbnail.get('height')),
703 'width': int_or_none(thumbnail.get('width')),
704 })
705 return thumbnails
706
707 @staticmethod
708 def extract_relative_time(relative_time_text):
709 """
710 Extracts a relative time from string and converts to dt object
711 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
712 """
713 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)
714 if mobj:
715 start = mobj.group('start')
716 if start:
717 return datetime_from_str(start)
718 try:
719 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
720 except ValueError:
721 return None
722
723 def _extract_time_text(self, renderer, *path_list):
724 text = self._get_text(renderer, *path_list) or ''
725 dt = self.extract_relative_time(text)
726 timestamp = None
727 if isinstance(dt, datetime.datetime):
728 timestamp = calendar.timegm(dt.timetuple())
729
730 if timestamp is None:
731 timestamp = (
732 unified_timestamp(text) or unified_timestamp(
733 self._search_regex(
734 (r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*on)?\s*(.+\d)', r'\w+[\s,\.-]*\w+[\s,\.-]+20\d{2}'),
735 text.lower(), 'time text', default=None)))
736
737 if text and timestamp is None:
738 self.report_warning('Cannot parse localized time text' + bug_reports_message(), only_once=True)
739 return timestamp, text
740
741 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
742 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
743 default_client='web'):
744 response = None
745 last_error = None
746 count = -1
747 retries = self.get_param('extractor_retries', 3)
748 if check_get_keys is None:
749 check_get_keys = []
750 while count < retries:
751 count += 1
752 if last_error:
753 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
754 try:
755 response = self._call_api(
756 ep=ep, fatal=True, headers=headers,
757 video_id=item_id, query=query,
758 context=self._extract_context(ytcfg, default_client),
759 api_key=self._extract_api_key(ytcfg, default_client),
760 api_hostname=api_hostname, default_client=default_client,
761 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
762 except ExtractorError as e:
763 if isinstance(e.cause, network_exceptions):
764 if isinstance(e.cause, compat_HTTPError):
765 first_bytes = e.cause.read(512)
766 if not is_html(first_bytes):
767 yt_error = try_get(
768 self._parse_json(
769 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
770 lambda x: x['error']['message'], compat_str)
771 if yt_error:
772 self._report_alerts([('ERROR', yt_error)], fatal=False)
773 # Downloading page may result in intermittent 5xx HTTP error
774 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
775 # We also want to catch all other network exceptions since errors in later pages can be troublesome
776 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
777 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
778 last_error = error_to_compat_str(e.cause or e.msg)
779 if count < retries:
780 continue
781 if fatal:
782 raise
783 else:
784 self.report_warning(error_to_compat_str(e))
785 return
786
787 else:
788 try:
789 self._extract_and_report_alerts(response, only_once=True)
790 except ExtractorError as e:
791 # YouTube servers may return errors we want to retry on in a 200 OK response
792 # See: https://github.com/yt-dlp/yt-dlp/issues/839
793 if 'unknown error' in e.msg.lower():
794 last_error = e.msg
795 continue
796 if fatal:
797 raise
798 self.report_warning(error_to_compat_str(e))
799 return
800 if not check_get_keys or dict_get(response, check_get_keys):
801 break
802 # Youtube sometimes sends incomplete data
803 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
804 last_error = 'Incomplete data received'
805 if count >= retries:
806 if fatal:
807 raise ExtractorError(last_error)
808 else:
809 self.report_warning(last_error)
810 return
811 return response
812
813 @staticmethod
814 def is_music_url(url):
815 return re.match(r'https?://music\.youtube\.com/', url) is not None
816
817 def _extract_video(self, renderer):
818 video_id = renderer.get('videoId')
819 title = self._get_text(renderer, 'title')
820 description = self._get_text(renderer, 'descriptionSnippet')
821 duration = parse_duration(self._get_text(
822 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
823 view_count = self._get_count(renderer, 'viewCountText')
824
825 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
826 channel_id = traverse_obj(
827 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'), expected_type=str, get_all=False)
828 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
829 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
830 overlay_style = traverse_obj(
831 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'), get_all=False, expected_type=str)
832 badges = self._extract_badges(renderer)
833 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
834
835 return {
836 '_type': 'url',
837 'ie_key': YoutubeIE.ie_key(),
838 'id': video_id,
839 'url': f'https://www.youtube.com/watch?v={video_id}',
840 'title': title,
841 'description': description,
842 'duration': duration,
843 'view_count': view_count,
844 'uploader': uploader,
845 'channel_id': channel_id,
846 'thumbnails': thumbnails,
847 # 'upload_date': strftime_or_none(timestamp, '%Y%m%d'),
848 'live_status': ('is_upcoming' if scheduled_timestamp is not None
849 else 'was_live' if 'streamed' in time_text.lower()
850 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
851 else None),
852 'release_timestamp': scheduled_timestamp,
853 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
854 }
855
856
857 class YoutubeIE(YoutubeBaseInfoExtractor):
858 IE_DESC = 'YouTube'
859 _VALID_URL = r"""(?x)^
860 (
861 (?:https?://|//) # http(s):// or protocol-independent URL
862 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
863 (?:www\.)?deturl\.com/www\.youtube\.com|
864 (?:www\.)?pwnyoutube\.com|
865 (?:www\.)?hooktube\.com|
866 (?:www\.)?yourepeat\.com|
867 tube\.majestyc\.net|
868 %(invidious)s|
869 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
870 (?:.*?\#/)? # handle anchor (#/) redirect urls
871 (?: # the various things that can precede the ID:
872 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
873 |(?: # or the v= param in all its forms
874 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
875 (?:\?|\#!?) # the params delimiter ? or # or #!
876 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
877 v=
878 )
879 ))
880 |(?:
881 youtu\.be| # just youtu.be/xxxx
882 vid\.plus| # or vid.plus/xxxx
883 zwearz\.com/watch| # or zwearz.com/watch/xxxx
884 %(invidious)s
885 )/
886 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
887 )
888 )? # all until now is optional -> you can pass the naked ID
889 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
890 (?(1).+)? # if we found the ID, everything can follow
891 (?:\#|$)""" % {
892 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
893 }
894 _PLAYER_INFO_RE = (
895 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
896 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
897 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
898 )
899 _formats = {
900 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
901 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
902 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
903 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
904 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
905 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
906 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
907 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
908 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
909 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
910 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
911 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
912 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
913 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
914 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
915 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
916 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
917 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
918
919
920 # 3D videos
921 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
922 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
923 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
924 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
925 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
926 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
927 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
928
929 # Apple HTTP Live Streaming
930 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
931 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
932 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
933 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
934 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
935 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
936 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
937 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
938
939 # DASH mp4 video
940 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
941 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
942 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
943 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
944 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
945 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
946 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
947 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
948 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
949 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
950 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
951 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
952
953 # Dash mp4 audio
954 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
955 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
956 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
957 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
958 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
959 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
960 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
961
962 # Dash webm
963 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
964 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
965 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
966 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
967 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
968 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
969 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
970 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
971 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
972 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
973 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
974 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
975 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
976 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
977 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
978 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
979 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
980 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
981 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
982 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
983 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
984 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
985
986 # Dash webm audio
987 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
988 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
989
990 # Dash webm audio with opus inside
991 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
992 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
993 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
994
995 # RTMP (unnamed)
996 '_rtmp': {'protocol': 'rtmp'},
997
998 # av01 video only formats sometimes served with "unknown" codecs
999 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1000 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1001 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1002 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1003 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1004 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1005 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1006 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1007 }
1008 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1009
1010 _GEO_BYPASS = False
1011
1012 IE_NAME = 'youtube'
1013 _TESTS = [
1014 {
1015 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1016 'info_dict': {
1017 'id': 'BaW_jenozKc',
1018 'ext': 'mp4',
1019 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1020 'uploader': 'Philipp Hagemeister',
1021 'uploader_id': 'phihag',
1022 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1023 'channel': 'Philipp Hagemeister',
1024 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1025 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1026 'upload_date': '20121002',
1027 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1028 'categories': ['Science & Technology'],
1029 'tags': ['youtube-dl'],
1030 'duration': 10,
1031 'view_count': int,
1032 'like_count': int,
1033 'availability': 'public',
1034 'playable_in_embed': True,
1035 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1036 'live_status': 'not_live',
1037 'age_limit': 0,
1038 'start_time': 1,
1039 'end_time': 9,
1040 'channel_follower_count': int
1041 }
1042 },
1043 {
1044 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1045 'note': 'Embed-only video (#1746)',
1046 'info_dict': {
1047 'id': 'yZIXLfi8CZQ',
1048 'ext': 'mp4',
1049 'upload_date': '20120608',
1050 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1051 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1052 'uploader': 'SET India',
1053 'uploader_id': 'setindia',
1054 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1055 'age_limit': 18,
1056 },
1057 'skip': 'Private video',
1058 },
1059 {
1060 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1061 'note': 'Use the first video ID in the URL',
1062 'info_dict': {
1063 'id': 'BaW_jenozKc',
1064 'ext': 'mp4',
1065 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1066 'uploader': 'Philipp Hagemeister',
1067 'uploader_id': 'phihag',
1068 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1069 'channel': 'Philipp Hagemeister',
1070 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1071 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1072 'upload_date': '20121002',
1073 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1074 'categories': ['Science & Technology'],
1075 'tags': ['youtube-dl'],
1076 'duration': 10,
1077 'view_count': int,
1078 'like_count': int,
1079 'availability': 'public',
1080 'playable_in_embed': True,
1081 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1082 'live_status': 'not_live',
1083 'age_limit': 0,
1084 'channel_follower_count': int
1085 },
1086 'params': {
1087 'skip_download': True,
1088 },
1089 },
1090 {
1091 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1092 'note': '256k DASH audio (format 141) via DASH manifest',
1093 'info_dict': {
1094 'id': 'a9LDPn-MO4I',
1095 'ext': 'm4a',
1096 'upload_date': '20121002',
1097 'uploader_id': '8KVIDEO',
1098 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1099 'description': '',
1100 'uploader': '8KVIDEO',
1101 'title': 'UHDTV TEST 8K VIDEO.mp4'
1102 },
1103 'params': {
1104 'youtube_include_dash_manifest': True,
1105 'format': '141',
1106 },
1107 'skip': 'format 141 not served anymore',
1108 },
1109 # DASH manifest with encrypted signature
1110 {
1111 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1112 'info_dict': {
1113 'id': 'IB3lcPjvWLA',
1114 'ext': 'm4a',
1115 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1116 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1117 'duration': 244,
1118 'uploader': 'AfrojackVEVO',
1119 'uploader_id': 'AfrojackVEVO',
1120 'upload_date': '20131011',
1121 'abr': 129.495,
1122 'like_count': int,
1123 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1124 'playable_in_embed': True,
1125 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1126 'view_count': int,
1127 'track': 'The Spark',
1128 'live_status': 'not_live',
1129 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1130 'channel': 'Afrojack',
1131 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1132 'tags': 'count:19',
1133 'availability': 'public',
1134 'categories': ['Music'],
1135 'age_limit': 0,
1136 'alt_title': 'The Spark',
1137 'channel_follower_count': int
1138 },
1139 'params': {
1140 'youtube_include_dash_manifest': True,
1141 'format': '141/bestaudio[ext=m4a]',
1142 },
1143 },
1144 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1145 {
1146 'note': 'Embed allowed age-gate video',
1147 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1148 'info_dict': {
1149 'id': 'HtVdAasjOgU',
1150 'ext': 'mp4',
1151 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1152 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1153 'duration': 142,
1154 'uploader': 'The Witcher',
1155 'uploader_id': 'WitcherGame',
1156 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1157 'upload_date': '20140605',
1158 'age_limit': 18,
1159 'categories': ['Gaming'],
1160 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1161 'availability': 'needs_auth',
1162 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1163 'like_count': int,
1164 'channel': 'The Witcher',
1165 'live_status': 'not_live',
1166 'tags': 'count:17',
1167 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1168 'playable_in_embed': True,
1169 'view_count': int,
1170 'channel_follower_count': int
1171 },
1172 },
1173 {
1174 'note': 'Age-gate video with embed allowed in public site',
1175 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1176 'info_dict': {
1177 'id': 'HsUATh_Nc2U',
1178 'ext': 'mp4',
1179 'title': 'Godzilla 2 (Official Video)',
1180 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1181 'upload_date': '20200408',
1182 'uploader_id': 'FlyingKitty900',
1183 'uploader': 'FlyingKitty',
1184 'age_limit': 18,
1185 'availability': 'needs_auth',
1186 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1187 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1188 'channel': 'FlyingKitty',
1189 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1190 'view_count': int,
1191 'categories': ['Entertainment'],
1192 'live_status': 'not_live',
1193 'tags': ['Flyingkitty', 'godzilla 2'],
1194 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1195 'like_count': int,
1196 'duration': 177,
1197 'playable_in_embed': True,
1198 'channel_follower_count': int
1199 },
1200 },
1201 {
1202 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1203 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1204 'info_dict': {
1205 'id': 'Tq92D6wQ1mg',
1206 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1207 'ext': 'mp4',
1208 'upload_date': '20191227',
1209 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1210 'uploader': 'Projekt Melody',
1211 'description': 'md5:17eccca93a786d51bc67646756894066',
1212 'age_limit': 18,
1213 'like_count': int,
1214 'availability': 'needs_auth',
1215 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1216 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1217 'view_count': int,
1218 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1219 'channel': 'Projekt Melody',
1220 'live_status': 'not_live',
1221 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1222 'playable_in_embed': True,
1223 'categories': ['Entertainment'],
1224 'duration': 106,
1225 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1226 'channel_follower_count': int
1227 },
1228 },
1229 {
1230 'note': 'Non-Agegated non-embeddable video',
1231 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1232 'info_dict': {
1233 'id': 'MeJVWBSsPAY',
1234 'ext': 'mp4',
1235 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1236 'uploader': 'Herr Lurik',
1237 'uploader_id': 'st3in234',
1238 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1239 'upload_date': '20130730',
1240 'track': 'Such mich find mich',
1241 'age_limit': 0,
1242 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1243 'like_count': int,
1244 'playable_in_embed': False,
1245 'creator': 'OOMPH!',
1246 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1247 'view_count': int,
1248 'alt_title': 'Such mich find mich',
1249 'duration': 210,
1250 'channel': 'Herr Lurik',
1251 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1252 'categories': ['Music'],
1253 'availability': 'public',
1254 'uploader_url': 'http://www.youtube.com/user/st3in234',
1255 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1256 'live_status': 'not_live',
1257 'artist': 'OOMPH!',
1258 'channel_follower_count': int
1259 },
1260 },
1261 {
1262 'note': 'Non-bypassable age-gated video',
1263 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1264 'only_matching': True,
1265 },
1266 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1267 # YouTube Red ad is not captured for creator
1268 {
1269 'url': '__2ABJjxzNo',
1270 'info_dict': {
1271 'id': '__2ABJjxzNo',
1272 'ext': 'mp4',
1273 'duration': 266,
1274 'upload_date': '20100430',
1275 'uploader_id': 'deadmau5',
1276 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1277 'creator': 'deadmau5',
1278 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1279 'uploader': 'deadmau5',
1280 'title': 'Deadmau5 - Some Chords (HD)',
1281 'alt_title': 'Some Chords',
1282 'availability': 'public',
1283 'tags': 'count:14',
1284 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1285 'view_count': int,
1286 'live_status': 'not_live',
1287 'channel': 'deadmau5',
1288 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1289 'like_count': int,
1290 'track': 'Some Chords',
1291 'artist': 'deadmau5',
1292 'playable_in_embed': True,
1293 'age_limit': 0,
1294 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1295 'categories': ['Music'],
1296 'album': 'Some Chords',
1297 'channel_follower_count': int
1298 },
1299 'expected_warnings': [
1300 'DASH manifest missing',
1301 ]
1302 },
1303 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1304 {
1305 'url': 'lqQg6PlCWgI',
1306 'info_dict': {
1307 'id': 'lqQg6PlCWgI',
1308 'ext': 'mp4',
1309 'duration': 6085,
1310 'upload_date': '20150827',
1311 'uploader_id': 'olympic',
1312 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1313 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1314 'uploader': 'Olympics',
1315 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1316 'like_count': int,
1317 'release_timestamp': 1343767800,
1318 'playable_in_embed': True,
1319 'categories': ['Sports'],
1320 'release_date': '20120731',
1321 'channel': 'Olympics',
1322 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1323 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1324 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1325 'age_limit': 0,
1326 'availability': 'public',
1327 'live_status': 'was_live',
1328 'view_count': int,
1329 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1330 'channel_follower_count': int
1331 },
1332 'params': {
1333 'skip_download': 'requires avconv',
1334 }
1335 },
1336 # Non-square pixels
1337 {
1338 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1339 'info_dict': {
1340 'id': '_b-2C3KPAM0',
1341 'ext': 'mp4',
1342 'stretched_ratio': 16 / 9.,
1343 'duration': 85,
1344 'upload_date': '20110310',
1345 'uploader_id': 'AllenMeow',
1346 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1347 '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',
1348 'uploader': '孫ᄋᄅ',
1349 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1350 'playable_in_embed': True,
1351 'channel': '孫ᄋᄅ',
1352 'age_limit': 0,
1353 'tags': 'count:11',
1354 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1355 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1356 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1357 'view_count': int,
1358 'categories': ['People & Blogs'],
1359 'like_count': int,
1360 'live_status': 'not_live',
1361 'availability': 'unlisted',
1362 'channel_follower_count': int
1363 },
1364 },
1365 # url_encoded_fmt_stream_map is empty string
1366 {
1367 'url': 'qEJwOuvDf7I',
1368 'info_dict': {
1369 'id': 'qEJwOuvDf7I',
1370 'ext': 'webm',
1371 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1372 'description': '',
1373 'upload_date': '20150404',
1374 'uploader_id': 'spbelect',
1375 'uploader': 'Наблюдатели Петербурга',
1376 },
1377 'params': {
1378 'skip_download': 'requires avconv',
1379 },
1380 'skip': 'This live event has ended.',
1381 },
1382 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1383 {
1384 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1385 'info_dict': {
1386 'id': 'FIl7x6_3R5Y',
1387 'ext': 'webm',
1388 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1389 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1390 'duration': 220,
1391 'upload_date': '20150625',
1392 'uploader_id': 'dorappi2000',
1393 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1394 'uploader': 'dorappi2000',
1395 'formats': 'mincount:31',
1396 },
1397 'skip': 'not actual anymore',
1398 },
1399 # DASH manifest with segment_list
1400 {
1401 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1402 'md5': '8ce563a1d667b599d21064e982ab9e31',
1403 'info_dict': {
1404 'id': 'CsmdDsKjzN8',
1405 'ext': 'mp4',
1406 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1407 'uploader': 'Airtek',
1408 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1409 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1410 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1411 },
1412 'params': {
1413 'youtube_include_dash_manifest': True,
1414 'format': '135', # bestvideo
1415 },
1416 'skip': 'This live event has ended.',
1417 },
1418 {
1419 # Multifeed videos (multiple cameras), URL is for Main Camera
1420 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1421 'info_dict': {
1422 'id': 'jvGDaLqkpTg',
1423 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1424 'description': 'md5:e03b909557865076822aa169218d6a5d',
1425 },
1426 'playlist': [{
1427 'info_dict': {
1428 'id': 'jvGDaLqkpTg',
1429 'ext': 'mp4',
1430 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1431 'description': 'md5:e03b909557865076822aa169218d6a5d',
1432 'duration': 10643,
1433 'upload_date': '20161111',
1434 'uploader': 'Team PGP',
1435 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1436 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1437 },
1438 }, {
1439 'info_dict': {
1440 'id': '3AKt1R1aDnw',
1441 'ext': 'mp4',
1442 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1443 'description': 'md5:e03b909557865076822aa169218d6a5d',
1444 'duration': 10991,
1445 'upload_date': '20161111',
1446 'uploader': 'Team PGP',
1447 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1448 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1449 },
1450 }, {
1451 'info_dict': {
1452 'id': 'RtAMM00gpVc',
1453 'ext': 'mp4',
1454 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1455 'description': 'md5:e03b909557865076822aa169218d6a5d',
1456 'duration': 10995,
1457 'upload_date': '20161111',
1458 'uploader': 'Team PGP',
1459 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1460 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1461 },
1462 }, {
1463 'info_dict': {
1464 'id': '6N2fdlP3C5U',
1465 'ext': 'mp4',
1466 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1467 'description': 'md5:e03b909557865076822aa169218d6a5d',
1468 'duration': 10990,
1469 'upload_date': '20161111',
1470 'uploader': 'Team PGP',
1471 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1472 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1473 },
1474 }],
1475 'params': {
1476 'skip_download': True,
1477 },
1478 'skip': 'Not multifeed anymore',
1479 },
1480 {
1481 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1482 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1483 'info_dict': {
1484 'id': 'gVfLd0zydlo',
1485 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1486 },
1487 'playlist_count': 2,
1488 'skip': 'Not multifeed anymore',
1489 },
1490 {
1491 'url': 'https://vid.plus/FlRa-iH7PGw',
1492 'only_matching': True,
1493 },
1494 {
1495 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1496 'only_matching': True,
1497 },
1498 {
1499 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1500 # Also tests cut-off URL expansion in video description (see
1501 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1502 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1503 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1504 'info_dict': {
1505 'id': 'lsguqyKfVQg',
1506 'ext': 'mp4',
1507 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1508 'alt_title': 'Dark Walk',
1509 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1510 'duration': 133,
1511 'upload_date': '20151119',
1512 'uploader_id': 'IronSoulElf',
1513 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1514 'uploader': 'IronSoulElf',
1515 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1516 'track': 'Dark Walk',
1517 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1518 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1519 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1520 'categories': ['Film & Animation'],
1521 'view_count': int,
1522 'live_status': 'not_live',
1523 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1524 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1525 'tags': 'count:13',
1526 'availability': 'public',
1527 'channel': 'IronSoulElf',
1528 'playable_in_embed': True,
1529 'like_count': int,
1530 'age_limit': 0,
1531 'channel_follower_count': int
1532 },
1533 'params': {
1534 'skip_download': True,
1535 },
1536 },
1537 {
1538 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1539 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1540 'only_matching': True,
1541 },
1542 {
1543 # Video with yt:stretch=17:0
1544 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1545 'info_dict': {
1546 'id': 'Q39EVAstoRM',
1547 'ext': 'mp4',
1548 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1549 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1550 'upload_date': '20151107',
1551 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1552 'uploader': 'CH GAMER DROID',
1553 },
1554 'params': {
1555 'skip_download': True,
1556 },
1557 'skip': 'This video does not exist.',
1558 },
1559 {
1560 # Video with incomplete 'yt:stretch=16:'
1561 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1562 'only_matching': True,
1563 },
1564 {
1565 # Video licensed under Creative Commons
1566 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1567 'info_dict': {
1568 'id': 'M4gD1WSo5mA',
1569 'ext': 'mp4',
1570 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1571 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1572 'duration': 721,
1573 'upload_date': '20150127',
1574 'uploader_id': 'BerkmanCenter',
1575 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1576 'uploader': 'The Berkman Klein Center for Internet & Society',
1577 'license': 'Creative Commons Attribution license (reuse allowed)',
1578 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1579 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1580 'like_count': int,
1581 'age_limit': 0,
1582 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1583 'channel': 'The Berkman Klein Center for Internet & Society',
1584 'availability': 'public',
1585 'view_count': int,
1586 'categories': ['Education'],
1587 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1588 'live_status': 'not_live',
1589 'playable_in_embed': True,
1590 'channel_follower_count': int
1591 },
1592 'params': {
1593 'skip_download': True,
1594 },
1595 },
1596 {
1597 # Channel-like uploader_url
1598 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1599 'info_dict': {
1600 'id': 'eQcmzGIKrzg',
1601 'ext': 'mp4',
1602 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1603 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1604 'duration': 4060,
1605 'upload_date': '20151119',
1606 'uploader': 'Bernie Sanders',
1607 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1608 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1609 'license': 'Creative Commons Attribution license (reuse allowed)',
1610 'playable_in_embed': True,
1611 'tags': 'count:12',
1612 'like_count': int,
1613 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1614 'age_limit': 0,
1615 'availability': 'public',
1616 'categories': ['News & Politics'],
1617 'channel': 'Bernie Sanders',
1618 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1619 'view_count': int,
1620 'live_status': 'not_live',
1621 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1622 'channel_follower_count': int
1623 },
1624 'params': {
1625 'skip_download': True,
1626 },
1627 },
1628 {
1629 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1630 'only_matching': True,
1631 },
1632 {
1633 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1634 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1635 'only_matching': True,
1636 },
1637 {
1638 # Rental video preview
1639 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1640 'info_dict': {
1641 'id': 'uGpuVWrhIzE',
1642 'ext': 'mp4',
1643 'title': 'Piku - Trailer',
1644 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1645 'upload_date': '20150811',
1646 'uploader': 'FlixMatrix',
1647 'uploader_id': 'FlixMatrixKaravan',
1648 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1649 'license': 'Standard YouTube License',
1650 },
1651 'params': {
1652 'skip_download': True,
1653 },
1654 'skip': 'This video is not available.',
1655 },
1656 {
1657 # YouTube Red video with episode data
1658 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1659 'info_dict': {
1660 'id': 'iqKdEhx-dD4',
1661 'ext': 'mp4',
1662 'title': 'Isolation - Mind Field (Ep 1)',
1663 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1664 'duration': 2085,
1665 'upload_date': '20170118',
1666 'uploader': 'Vsauce',
1667 'uploader_id': 'Vsauce',
1668 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1669 'series': 'Mind Field',
1670 'season_number': 1,
1671 'episode_number': 1,
1672 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1673 'tags': 'count:12',
1674 'view_count': int,
1675 'availability': 'public',
1676 'age_limit': 0,
1677 'channel': 'Vsauce',
1678 'episode': 'Episode 1',
1679 'categories': ['Entertainment'],
1680 'season': 'Season 1',
1681 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1682 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1683 'like_count': int,
1684 'playable_in_embed': True,
1685 'live_status': 'not_live',
1686 'channel_follower_count': int
1687 },
1688 'params': {
1689 'skip_download': True,
1690 },
1691 'expected_warnings': [
1692 'Skipping DASH manifest',
1693 ],
1694 },
1695 {
1696 # The following content has been identified by the YouTube community
1697 # as inappropriate or offensive to some audiences.
1698 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1699 'info_dict': {
1700 'id': '6SJNVb0GnPI',
1701 'ext': 'mp4',
1702 'title': 'Race Differences in Intelligence',
1703 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1704 'duration': 965,
1705 'upload_date': '20140124',
1706 'uploader': 'New Century Foundation',
1707 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1708 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1709 },
1710 'params': {
1711 'skip_download': True,
1712 },
1713 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1714 },
1715 {
1716 # itag 212
1717 'url': '1t24XAntNCY',
1718 'only_matching': True,
1719 },
1720 {
1721 # geo restricted to JP
1722 'url': 'sJL6WA-aGkQ',
1723 'only_matching': True,
1724 },
1725 {
1726 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1727 'only_matching': True,
1728 },
1729 {
1730 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1731 'only_matching': True,
1732 },
1733 {
1734 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1735 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1736 'only_matching': True,
1737 },
1738 {
1739 # DRM protected
1740 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1741 'only_matching': True,
1742 },
1743 {
1744 # Video with unsupported adaptive stream type formats
1745 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1746 'info_dict': {
1747 'id': 'Z4Vy8R84T1U',
1748 'ext': 'mp4',
1749 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1750 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1751 'duration': 433,
1752 'upload_date': '20130923',
1753 'uploader': 'Amelia Putri Harwita',
1754 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1755 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1756 'formats': 'maxcount:10',
1757 },
1758 'params': {
1759 'skip_download': True,
1760 'youtube_include_dash_manifest': False,
1761 },
1762 'skip': 'not actual anymore',
1763 },
1764 {
1765 # Youtube Music Auto-generated description
1766 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1767 'info_dict': {
1768 'id': 'MgNrAu2pzNs',
1769 'ext': 'mp4',
1770 'title': 'Voyeur Girl',
1771 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1772 'upload_date': '20190312',
1773 'uploader': 'Stephen - Topic',
1774 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1775 'artist': 'Stephen',
1776 'track': 'Voyeur Girl',
1777 'album': 'it\'s too much love to know my dear',
1778 'release_date': '20190313',
1779 'release_year': 2019,
1780 'alt_title': 'Voyeur Girl',
1781 'view_count': int,
1782 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1783 'playable_in_embed': True,
1784 'like_count': int,
1785 'categories': ['Music'],
1786 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1787 'channel': 'Stephen',
1788 'availability': 'public',
1789 'creator': 'Stephen',
1790 'duration': 169,
1791 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1792 'age_limit': 0,
1793 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1794 'tags': 'count:11',
1795 'live_status': 'not_live',
1796 'channel_follower_count': int
1797 },
1798 'params': {
1799 'skip_download': True,
1800 },
1801 },
1802 {
1803 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1804 'only_matching': True,
1805 },
1806 {
1807 # invalid -> valid video id redirection
1808 'url': 'DJztXj2GPfl',
1809 'info_dict': {
1810 'id': 'DJztXj2GPfk',
1811 'ext': 'mp4',
1812 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1813 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1814 'upload_date': '20090125',
1815 'uploader': 'Prochorowka',
1816 'uploader_id': 'Prochorowka',
1817 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1818 'artist': 'Panjabi MC',
1819 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1820 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1821 },
1822 'params': {
1823 'skip_download': True,
1824 },
1825 'skip': 'Video unavailable',
1826 },
1827 {
1828 # empty description results in an empty string
1829 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1830 'info_dict': {
1831 'id': 'x41yOUIvK2k',
1832 'ext': 'mp4',
1833 'title': 'IMG 3456',
1834 'description': '',
1835 'upload_date': '20170613',
1836 'uploader_id': 'ElevageOrVert',
1837 'uploader': 'ElevageOrVert',
1838 'view_count': int,
1839 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1840 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1841 'like_count': int,
1842 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1843 'tags': [],
1844 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1845 'availability': 'public',
1846 'age_limit': 0,
1847 'categories': ['Pets & Animals'],
1848 'duration': 7,
1849 'playable_in_embed': True,
1850 'live_status': 'not_live',
1851 'channel': 'ElevageOrVert',
1852 'channel_follower_count': int
1853 },
1854 'params': {
1855 'skip_download': True,
1856 },
1857 },
1858 {
1859 # with '};' inside yt initial data (see [1])
1860 # see [2] for an example with '};' inside ytInitialPlayerResponse
1861 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1862 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1863 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1864 'info_dict': {
1865 'id': 'CHqg6qOn4no',
1866 'ext': 'mp4',
1867 'title': 'Part 77 Sort a list of simple types in c#',
1868 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1869 'upload_date': '20130831',
1870 'uploader_id': 'kudvenkat',
1871 'uploader': 'kudvenkat',
1872 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1873 'like_count': int,
1874 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1875 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1876 'live_status': 'not_live',
1877 'categories': ['Education'],
1878 'availability': 'public',
1879 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1880 'tags': 'count:12',
1881 'playable_in_embed': True,
1882 'age_limit': 0,
1883 'view_count': int,
1884 'duration': 522,
1885 'channel': 'kudvenkat',
1886 'channel_follower_count': int
1887 },
1888 'params': {
1889 'skip_download': True,
1890 },
1891 },
1892 {
1893 # another example of '};' in ytInitialData
1894 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1895 'only_matching': True,
1896 },
1897 {
1898 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1899 'only_matching': True,
1900 },
1901 {
1902 # https://github.com/ytdl-org/youtube-dl/pull/28094
1903 'url': 'OtqTfy26tG0',
1904 'info_dict': {
1905 'id': 'OtqTfy26tG0',
1906 'ext': 'mp4',
1907 'title': 'Burn Out',
1908 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1909 'upload_date': '20141120',
1910 'uploader': 'The Cinematic Orchestra - Topic',
1911 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1912 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1913 'artist': 'The Cinematic Orchestra',
1914 'track': 'Burn Out',
1915 'album': 'Every Day',
1916 'like_count': int,
1917 'live_status': 'not_live',
1918 'alt_title': 'Burn Out',
1919 'duration': 614,
1920 'age_limit': 0,
1921 'view_count': int,
1922 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1923 'creator': 'The Cinematic Orchestra',
1924 'channel': 'The Cinematic Orchestra',
1925 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1926 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1927 'availability': 'public',
1928 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1929 'categories': ['Music'],
1930 'playable_in_embed': True,
1931 'channel_follower_count': int
1932 },
1933 'params': {
1934 'skip_download': True,
1935 },
1936 },
1937 {
1938 # controversial video, only works with bpctr when authenticated with cookies
1939 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1940 'only_matching': True,
1941 },
1942 {
1943 # controversial video, requires bpctr/contentCheckOk
1944 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1945 'info_dict': {
1946 'id': 'SZJvDhaSDnc',
1947 'ext': 'mp4',
1948 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1949 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1950 'uploader': 'CBS Mornings',
1951 'uploader_id': 'CBSThisMorning',
1952 'upload_date': '20140716',
1953 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
1954 'duration': 170,
1955 'categories': ['News & Politics'],
1956 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
1957 'view_count': int,
1958 'channel': 'CBS Mornings',
1959 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
1960 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
1961 'age_limit': 18,
1962 'availability': 'needs_auth',
1963 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
1964 'like_count': int,
1965 'live_status': 'not_live',
1966 'playable_in_embed': True,
1967 'channel_follower_count': int
1968 }
1969 },
1970 {
1971 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
1972 'url': 'cBvYw8_A0vQ',
1973 'info_dict': {
1974 'id': 'cBvYw8_A0vQ',
1975 'ext': 'mp4',
1976 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
1977 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
1978 'upload_date': '20201120',
1979 'uploader': 'Walk around Japan',
1980 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
1981 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
1982 'duration': 1456,
1983 'categories': ['Travel & Events'],
1984 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
1985 'view_count': int,
1986 'channel': 'Walk around Japan',
1987 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
1988 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
1989 'age_limit': 0,
1990 'availability': 'public',
1991 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
1992 'live_status': 'not_live',
1993 'playable_in_embed': True,
1994 'channel_follower_count': int
1995 },
1996 'params': {
1997 'skip_download': True,
1998 },
1999 }, {
2000 # Has multiple audio streams
2001 'url': 'WaOKSUlf4TM',
2002 'only_matching': True
2003 }, {
2004 # Requires Premium: has format 141 when requested using YTM url
2005 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2006 'only_matching': True
2007 }, {
2008 # multiple subtitles with same lang_code
2009 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2010 'only_matching': True,
2011 }, {
2012 # Force use android client fallback
2013 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2014 'info_dict': {
2015 'id': 'YOelRv7fMxY',
2016 'title': 'DIGGING A SECRET TUNNEL Part 1',
2017 'ext': '3gp',
2018 'upload_date': '20210624',
2019 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2020 'uploader': 'colinfurze',
2021 'uploader_id': 'colinfurze',
2022 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2023 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2024 'duration': 596,
2025 'categories': ['Entertainment'],
2026 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2027 'view_count': int,
2028 'channel': 'colinfurze',
2029 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2030 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2031 'age_limit': 0,
2032 'availability': 'public',
2033 'like_count': int,
2034 'live_status': 'not_live',
2035 'playable_in_embed': True,
2036 'channel_follower_count': int
2037 },
2038 'params': {
2039 'format': '17', # 3gp format available on android
2040 'extractor_args': {'youtube': {'player_client': ['android']}},
2041 },
2042 },
2043 {
2044 # Skip download of additional client configs (remix client config in this case)
2045 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2046 'only_matching': True,
2047 'params': {
2048 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2049 },
2050 }, {
2051 # shorts
2052 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2053 'only_matching': True,
2054 }, {
2055 'note': 'Storyboards',
2056 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2057 'info_dict': {
2058 'id': '5KLPxDtMqe8',
2059 'ext': 'mhtml',
2060 'format_id': 'sb0',
2061 'title': 'Your Brain is Plastic',
2062 'uploader_id': 'scishow',
2063 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2064 'upload_date': '20140324',
2065 'uploader': 'SciShow',
2066 'like_count': int,
2067 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2068 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2069 'view_count': int,
2070 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2071 'playable_in_embed': True,
2072 'tags': 'count:12',
2073 'uploader_url': 'http://www.youtube.com/user/scishow',
2074 'availability': 'public',
2075 'channel': 'SciShow',
2076 'live_status': 'not_live',
2077 'duration': 248,
2078 'categories': ['Education'],
2079 'age_limit': 0,
2080 'channel_follower_count': int
2081 }, 'params': {'format': 'mhtml', 'skip_download': True}
2082 }
2083 ]
2084
2085 @classmethod
2086 def suitable(cls, url):
2087 from ..utils import parse_qs
2088
2089 qs = parse_qs(url)
2090 if qs.get('list', [None])[0]:
2091 return False
2092 return super(YoutubeIE, cls).suitable(url)
2093
2094 def __init__(self, *args, **kwargs):
2095 super(YoutubeIE, self).__init__(*args, **kwargs)
2096 self._code_cache = {}
2097 self._player_cache = {}
2098
2099 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2100 lock = threading.Lock()
2101
2102 is_live = True
2103 start_time = time.time()
2104 formats = [f for f in formats if f.get('is_from_start')]
2105
2106 def refetch_manifest(format_id, delay):
2107 nonlocal formats, start_time, is_live
2108 if time.time() <= start_time + delay:
2109 return
2110
2111 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2112 video_details = traverse_obj(
2113 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2114 microformats = traverse_obj(
2115 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2116 expected_type=dict, default=[])
2117 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2118 start_time = time.time()
2119
2120 def mpd_feed(format_id, delay):
2121 """
2122 @returns (manifest_url, manifest_stream_number, is_live) or None
2123 """
2124 with lock:
2125 refetch_manifest(format_id, delay)
2126
2127 f = next((f for f in formats if f['format_id'] == format_id), None)
2128 if not f:
2129 if not is_live:
2130 self.to_screen(f'{video_id}: Video is no longer live')
2131 else:
2132 self.report_warning(
2133 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2134 return None
2135 return f['manifest_url'], f['manifest_stream_number'], is_live
2136
2137 for f in formats:
2138 f['protocol'] = 'http_dash_segments_generator'
2139 f['fragments'] = functools.partial(
2140 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2141
2142 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2143 FETCH_SPAN, MAX_DURATION = 5, 432000
2144
2145 mpd_url, stream_number, is_live = None, None, True
2146
2147 begin_index = 0
2148 download_start_time = ctx.get('start') or time.time()
2149
2150 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2151 if lack_early_segments:
2152 self.report_warning(bug_reports_message(
2153 'Starting download from the last 120 hours of the live stream since '
2154 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2155 lack_early_segments = True
2156
2157 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2158 fragments, fragment_base_url = None, None
2159
2160 def _extract_sequence_from_mpd(refresh_sequence):
2161 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2162 # Obtain from MPD's maximum seq value
2163 old_mpd_url = mpd_url
2164 last_error = ctx.pop('last_error', None)
2165 expire_fast = last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2166 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2167 or (mpd_url, stream_number, False))
2168 if not refresh_sequence:
2169 if expire_fast and not is_live:
2170 return False, last_seq
2171 elif old_mpd_url == mpd_url:
2172 return True, last_seq
2173 try:
2174 fmts, _ = self._extract_mpd_formats_and_subtitles(
2175 mpd_url, None, note=False, errnote=False, fatal=False)
2176 except ExtractorError:
2177 fmts = None
2178 if not fmts:
2179 no_fragment_score += 1
2180 return False, last_seq
2181 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2182 fragments = fmt_info['fragments']
2183 fragment_base_url = fmt_info['fragment_base_url']
2184 assert fragment_base_url
2185
2186 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2187 return True, _last_seq
2188
2189 while is_live:
2190 fetch_time = time.time()
2191 if no_fragment_score > 30:
2192 return
2193 if last_segment_url:
2194 # Obtain from "X-Head-Seqnum" header value from each segment
2195 try:
2196 urlh = self._request_webpage(
2197 last_segment_url, None, note=False, errnote=False, fatal=False)
2198 except ExtractorError:
2199 urlh = None
2200 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2201 if last_seq is None:
2202 no_fragment_score += 1
2203 last_segment_url = None
2204 continue
2205 else:
2206 should_continue, last_seq = _extract_sequence_from_mpd(True)
2207 if not should_continue:
2208 continue
2209
2210 if known_idx > last_seq:
2211 last_segment_url = None
2212 continue
2213
2214 last_seq += 1
2215
2216 if begin_index < 0 and known_idx < 0:
2217 # skip from the start when it's negative value
2218 known_idx = last_seq + begin_index
2219 if lack_early_segments:
2220 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2221 try:
2222 for idx in range(known_idx, last_seq):
2223 # do not update sequence here or you'll get skipped some part of it
2224 should_continue, _ = _extract_sequence_from_mpd(False)
2225 if not should_continue:
2226 known_idx = idx - 1
2227 raise ExtractorError('breaking out of outer loop')
2228 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2229 yield {
2230 'url': last_segment_url,
2231 }
2232 if known_idx == last_seq:
2233 no_fragment_score += 5
2234 else:
2235 no_fragment_score = 0
2236 known_idx = last_seq
2237 except ExtractorError:
2238 continue
2239
2240 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2241
2242 def _extract_player_url(self, *ytcfgs, webpage=None):
2243 player_url = traverse_obj(
2244 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2245 get_all=False, expected_type=compat_str)
2246 if not player_url:
2247 return
2248 if player_url.startswith('//'):
2249 player_url = 'https:' + player_url
2250 elif not re.match(r'https?://', player_url):
2251 player_url = compat_urlparse.urljoin(
2252 'https://www.youtube.com', player_url)
2253 return player_url
2254
2255 def _download_player_url(self, video_id, fatal=False):
2256 res = self._download_webpage(
2257 'https://www.youtube.com/iframe_api',
2258 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2259 if res:
2260 player_version = self._search_regex(
2261 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2262 if player_version:
2263 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2264
2265 def _signature_cache_id(self, example_sig):
2266 """ Return a string representation of a signature """
2267 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2268
2269 @classmethod
2270 def _extract_player_info(cls, player_url):
2271 for player_re in cls._PLAYER_INFO_RE:
2272 id_m = re.search(player_re, player_url)
2273 if id_m:
2274 break
2275 else:
2276 raise ExtractorError('Cannot identify player %r' % player_url)
2277 return id_m.group('id')
2278
2279 def _load_player(self, video_id, player_url, fatal=True):
2280 player_id = self._extract_player_info(player_url)
2281 if player_id not in self._code_cache:
2282 code = self._download_webpage(
2283 player_url, video_id, fatal=fatal,
2284 note='Downloading player ' + player_id,
2285 errnote='Download of %s failed' % player_url)
2286 if code:
2287 self._code_cache[player_id] = code
2288 return self._code_cache.get(player_id)
2289
2290 def _extract_signature_function(self, video_id, player_url, example_sig):
2291 player_id = self._extract_player_info(player_url)
2292
2293 # Read from filesystem cache
2294 func_id = 'js_%s_%s' % (
2295 player_id, self._signature_cache_id(example_sig))
2296 assert os.path.basename(func_id) == func_id
2297
2298 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2299 if cache_spec is not None:
2300 return lambda s: ''.join(s[i] for i in cache_spec)
2301
2302 code = self._load_player(video_id, player_url)
2303 if code:
2304 res = self._parse_sig_js(code)
2305
2306 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2307 cache_res = res(test_string)
2308 cache_spec = [ord(c) for c in cache_res]
2309
2310 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2311 return res
2312
2313 def _print_sig_code(self, func, example_sig):
2314 if not self.get_param('youtube_print_sig_code'):
2315 return
2316
2317 def gen_sig_code(idxs):
2318 def _genslice(start, end, step):
2319 starts = '' if start == 0 else str(start)
2320 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2321 steps = '' if step == 1 else (':%d' % step)
2322 return 's[%s%s%s]' % (starts, ends, steps)
2323
2324 step = None
2325 # Quelch pyflakes warnings - start will be set when step is set
2326 start = '(Never used)'
2327 for i, prev in zip(idxs[1:], idxs[:-1]):
2328 if step is not None:
2329 if i - prev == step:
2330 continue
2331 yield _genslice(start, prev, step)
2332 step = None
2333 continue
2334 if i - prev in [-1, 1]:
2335 step = i - prev
2336 start = prev
2337 continue
2338 else:
2339 yield 's[%d]' % prev
2340 if step is None:
2341 yield 's[%d]' % i
2342 else:
2343 yield _genslice(start, i, step)
2344
2345 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2346 cache_res = func(test_string)
2347 cache_spec = [ord(c) for c in cache_res]
2348 expr_code = ' + '.join(gen_sig_code(cache_spec))
2349 signature_id_tuple = '(%s)' % (
2350 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2351 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2352 ' return %s\n') % (signature_id_tuple, expr_code)
2353 self.to_screen('Extracted signature function:\n' + code)
2354
2355 def _parse_sig_js(self, jscode):
2356 funcname = self._search_regex(
2357 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2358 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2359 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2360 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2361 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}\(a,\d+\)',
2362 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\(\s*""\s*\)',
2363 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2364 # Obsolete patterns
2365 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2366 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2367 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2368 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2369 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2370 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2371 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2372 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2373 jscode, 'Initial JS player signature function name', group='sig')
2374
2375 jsi = JSInterpreter(jscode)
2376 initial_function = jsi.extract_function(funcname)
2377 return lambda s: initial_function([s])
2378
2379 def _decrypt_signature(self, s, video_id, player_url):
2380 """Turn the encrypted s field into a working signature"""
2381
2382 if player_url is None:
2383 raise ExtractorError('Cannot decrypt signature without player_url')
2384
2385 try:
2386 player_id = (player_url, self._signature_cache_id(s))
2387 if player_id not in self._player_cache:
2388 func = self._extract_signature_function(
2389 video_id, player_url, s
2390 )
2391 self._player_cache[player_id] = func
2392 func = self._player_cache[player_id]
2393 self._print_sig_code(func, s)
2394 return func(s)
2395 except Exception as e:
2396 raise ExtractorError('Signature extraction failed: ' + traceback.format_exc(), cause=e)
2397
2398 def _decrypt_nsig(self, s, video_id, player_url):
2399 """Turn the encrypted n field into a working signature"""
2400 if player_url is None:
2401 raise ExtractorError('Cannot decrypt nsig without player_url')
2402 if player_url.startswith('//'):
2403 player_url = 'https:' + player_url
2404 elif not re.match(r'https?://', player_url):
2405 player_url = compat_urlparse.urljoin(
2406 'https://www.youtube.com', player_url)
2407
2408 sig_id = ('nsig_value', s)
2409 if sig_id in self._player_cache:
2410 return self._player_cache[sig_id]
2411
2412 try:
2413 player_id = ('nsig', player_url)
2414 if player_id not in self._player_cache:
2415 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2416 func = self._player_cache[player_id]
2417 self._player_cache[sig_id] = func(s)
2418 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2419 return self._player_cache[sig_id]
2420 except Exception as e:
2421 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2422
2423 def _extract_n_function_name(self, jscode):
2424 nfunc, idx = self._search_regex(
2425 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]{3})(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2426 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2427 if not idx:
2428 return nfunc
2429 return json.loads(js_to_json(self._search_regex(
2430 rf'var {nfunc}\s*=\s*(\[.+?\]);', jscode,
2431 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2432
2433 def _extract_n_function(self, video_id, player_url):
2434 player_id = self._extract_player_info(player_url)
2435 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2436
2437 if func_code:
2438 jsi = JSInterpreter(func_code)
2439 else:
2440 jscode = self._load_player(video_id, player_url)
2441 funcname = self._extract_n_function_name(jscode)
2442 jsi = JSInterpreter(jscode)
2443 func_code = jsi.extract_function_code(funcname)
2444 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2445
2446 if self.get_param('youtube_print_sig_code'):
2447 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2448
2449 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2450
2451 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2452 """
2453 Extract signatureTimestamp (sts)
2454 Required to tell API what sig/player version is in use.
2455 """
2456 sts = None
2457 if isinstance(ytcfg, dict):
2458 sts = int_or_none(ytcfg.get('STS'))
2459
2460 if not sts:
2461 # Attempt to extract from player
2462 if player_url is None:
2463 error_msg = 'Cannot extract signature timestamp without player_url.'
2464 if fatal:
2465 raise ExtractorError(error_msg)
2466 self.report_warning(error_msg)
2467 return
2468 code = self._load_player(video_id, player_url, fatal=fatal)
2469 if code:
2470 sts = int_or_none(self._search_regex(
2471 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2472 'JS player signature timestamp', group='sts', fatal=fatal))
2473 return sts
2474
2475 def _mark_watched(self, video_id, player_responses):
2476 playback_url = get_first(
2477 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2478 expected_type=url_or_none)
2479 if not playback_url:
2480 self.report_warning('Unable to mark watched')
2481 return
2482 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2483 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2484
2485 # cpn generation algorithm is reverse engineered from base.js.
2486 # In fact it works even with dummy cpn.
2487 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2488 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
2489
2490 qs.update({
2491 'ver': ['2'],
2492 'cpn': [cpn],
2493 })
2494 playback_url = compat_urlparse.urlunparse(
2495 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2496
2497 self._download_webpage(
2498 playback_url, video_id, 'Marking watched',
2499 'Unable to mark watched', fatal=False)
2500
2501 @staticmethod
2502 def _extract_urls(webpage):
2503 # Embedded YouTube player
2504 entries = [
2505 unescapeHTML(mobj.group('url'))
2506 for mobj in re.finditer(r'''(?x)
2507 (?:
2508 <iframe[^>]+?src=|
2509 data-video-url=|
2510 <embed[^>]+?src=|
2511 embedSWF\(?:\s*|
2512 <object[^>]+data=|
2513 new\s+SWFObject\(
2514 )
2515 (["\'])
2516 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2517 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2518 \1''', webpage)]
2519
2520 # lazyYT YouTube embed
2521 entries.extend(list(map(
2522 unescapeHTML,
2523 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2524
2525 # Wordpress "YouTube Video Importer" plugin
2526 matches = re.findall(r'''(?x)<div[^>]+
2527 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2528 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2529 entries.extend(m[-1] for m in matches)
2530
2531 return entries
2532
2533 @staticmethod
2534 def _extract_url(webpage):
2535 urls = YoutubeIE._extract_urls(webpage)
2536 return urls[0] if urls else None
2537
2538 @classmethod
2539 def extract_id(cls, url):
2540 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2541 if mobj is None:
2542 raise ExtractorError('Invalid URL: %s' % url)
2543 return mobj.group('id')
2544
2545 def _extract_chapters_from_json(self, data, duration):
2546 chapter_list = traverse_obj(
2547 data, (
2548 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2549 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2550 ), expected_type=list)
2551
2552 return self._extract_chapters(
2553 chapter_list,
2554 chapter_time=lambda chapter: float_or_none(
2555 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2556 chapter_title=lambda chapter: traverse_obj(
2557 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2558 duration=duration)
2559
2560 def _extract_chapters_from_engagement_panel(self, data, duration):
2561 content_list = traverse_obj(
2562 data,
2563 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2564 expected_type=list, default=[])
2565 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2566 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2567
2568 return next((
2569 filter(None, (
2570 self._extract_chapters(
2571 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2572 chapter_time, chapter_title, duration)
2573 for contents in content_list
2574 ))), [])
2575
2576 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2577 chapters = []
2578 last_chapter = {'start_time': 0}
2579 for idx, chapter in enumerate(chapter_list or []):
2580 title = chapter_title(chapter)
2581 start_time = chapter_time(chapter)
2582 if start_time is None:
2583 continue
2584 last_chapter['end_time'] = start_time
2585 if start_time < last_chapter['start_time']:
2586 if idx == 1:
2587 chapters.pop()
2588 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2589 else:
2590 self.report_warning(f'Invalid start time for chapter "{title}"')
2591 continue
2592 last_chapter = {'start_time': start_time, 'title': title}
2593 chapters.append(last_chapter)
2594 last_chapter['end_time'] = duration
2595 return chapters
2596
2597 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2598 return self._parse_json(self._search_regex(
2599 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
2600 regex), webpage, name, default='{}'), video_id, fatal=False)
2601
2602 def _extract_comment(self, comment_renderer, parent=None):
2603 comment_id = comment_renderer.get('commentId')
2604 if not comment_id:
2605 return
2606
2607 text = self._get_text(comment_renderer, 'contentText')
2608
2609 # note: timestamp is an estimate calculated from the current time and time_text
2610 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2611 author = self._get_text(comment_renderer, 'authorText')
2612 author_id = try_get(comment_renderer,
2613 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2614
2615 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2616 lambda x: x['likeCount']), compat_str)) or 0
2617 author_thumbnail = try_get(comment_renderer,
2618 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2619
2620 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2621 is_favorited = 'creatorHeart' in (try_get(
2622 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2623 return {
2624 'id': comment_id,
2625 'text': text,
2626 'timestamp': timestamp,
2627 'time_text': time_text,
2628 'like_count': votes,
2629 'is_favorited': is_favorited,
2630 'author': author,
2631 'author_id': author_id,
2632 'author_thumbnail': author_thumbnail,
2633 'author_is_uploader': author_is_uploader,
2634 'parent': parent or 'root'
2635 }
2636
2637 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2638
2639 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2640
2641 def extract_header(contents):
2642 _continuation = None
2643 for content in contents:
2644 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2645 expected_comment_count = self._get_count(
2646 comments_header_renderer, 'countText', 'commentsCount')
2647
2648 if expected_comment_count:
2649 tracker['est_total'] = expected_comment_count
2650 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2651 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2652
2653 sort_menu_item = try_get(
2654 comments_header_renderer,
2655 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2656 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2657
2658 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2659 if not _continuation:
2660 continue
2661
2662 sort_text = str_or_none(sort_menu_item.get('title'))
2663 if not sort_text:
2664 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2665 self.to_screen('Sorting comments by %s' % sort_text.lower())
2666 break
2667 return _continuation
2668
2669 def extract_thread(contents):
2670 if not parent:
2671 tracker['current_page_thread'] = 0
2672 for content in contents:
2673 if not parent and tracker['total_parent_comments'] >= max_parents:
2674 yield
2675 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2676 comment_renderer = get_first(
2677 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2678 expected_type=dict, default={})
2679
2680 comment = self._extract_comment(comment_renderer, parent)
2681 if not comment:
2682 continue
2683
2684 tracker['running_total'] += 1
2685 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2686 yield comment
2687
2688 # Attempt to get the replies
2689 comment_replies_renderer = try_get(
2690 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2691
2692 if comment_replies_renderer:
2693 tracker['current_page_thread'] += 1
2694 comment_entries_iter = self._comment_entries(
2695 comment_replies_renderer, ytcfg, video_id,
2696 parent=comment.get('id'), tracker=tracker)
2697 for reply_comment in itertools.islice(comment_entries_iter, min(max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments']))):
2698 yield reply_comment
2699
2700 # Keeps track of counts across recursive calls
2701 if not tracker:
2702 tracker = dict(
2703 running_total=0,
2704 est_total=0,
2705 current_page_thread=0,
2706 total_parent_comments=0,
2707 total_reply_comments=0)
2708
2709 # TODO: Deprecated
2710 # YouTube comments have a max depth of 2
2711 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2712 if max_depth:
2713 self._downloader.deprecation_warning(
2714 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2715 if max_depth == 1 and parent:
2716 return
2717
2718 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2719 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2720
2721 continuation = self._extract_continuation(root_continuation_data)
2722 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2723 if message and not parent:
2724 self.report_warning(message, video_id=video_id)
2725
2726 response = None
2727 is_first_continuation = parent is None
2728
2729 for page_num in itertools.count(0):
2730 if not continuation:
2731 break
2732 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2733 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2734 if page_num == 0:
2735 if is_first_continuation:
2736 note_prefix = 'Downloading comment section API JSON'
2737 else:
2738 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2739 tracker['current_page_thread'], comment_prog_str)
2740 else:
2741 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2742 ' ' if parent else '', ' replies' if parent else '',
2743 page_num, comment_prog_str)
2744
2745 response = self._extract_response(
2746 item_id=None, query=continuation,
2747 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2748 check_get_keys='onResponseReceivedEndpoints')
2749
2750 continuation_contents = traverse_obj(
2751 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2752
2753 continuation = None
2754 for continuation_section in continuation_contents:
2755 continuation_items = traverse_obj(
2756 continuation_section,
2757 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2758 get_all=False, expected_type=list) or []
2759 if is_first_continuation:
2760 continuation = extract_header(continuation_items)
2761 is_first_continuation = False
2762 if continuation:
2763 break
2764 continue
2765
2766 for entry in extract_thread(continuation_items):
2767 if not entry:
2768 return
2769 yield entry
2770 continuation = self._extract_continuation({'contents': continuation_items})
2771 if continuation:
2772 break
2773
2774 def _get_comments(self, ytcfg, video_id, contents, webpage):
2775 """Entry for comment extraction"""
2776 def _real_comment_extract(contents):
2777 renderer = next((
2778 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2779 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2780 yield from self._comment_entries(renderer, ytcfg, video_id)
2781
2782 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2783 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2784
2785 @staticmethod
2786 def _get_checkok_params():
2787 return {'contentCheckOk': True, 'racyCheckOk': True}
2788
2789 @classmethod
2790 def _generate_player_context(cls, sts=None):
2791 context = {
2792 'html5Preference': 'HTML5_PREF_WANTS',
2793 }
2794 if sts is not None:
2795 context['signatureTimestamp'] = sts
2796 return {
2797 'playbackContext': {
2798 'contentPlaybackContext': context
2799 },
2800 **cls._get_checkok_params()
2801 }
2802
2803 @staticmethod
2804 def _is_agegated(player_response):
2805 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2806 return True
2807
2808 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2809 AGE_GATE_REASONS = (
2810 'confirm your age', 'age-restricted', 'inappropriate', # reason
2811 'age_verification_required', 'age_check_required', # status
2812 )
2813 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2814
2815 @staticmethod
2816 def _is_unplayable(player_response):
2817 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2818
2819 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2820
2821 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2822 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2823 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2824 headers = self.generate_api_headers(
2825 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
2826
2827 yt_query = {'videoId': video_id}
2828 yt_query.update(self._generate_player_context(sts))
2829 return self._extract_response(
2830 item_id=video_id, ep='player', query=yt_query,
2831 ytcfg=player_ytcfg, headers=headers, fatal=True,
2832 default_client=client,
2833 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
2834 ) or None
2835
2836 def _get_requested_clients(self, url, smuggled_data):
2837 requested_clients = []
2838 default = ['android', 'web']
2839 allowed_clients = sorted(
2840 [client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'],
2841 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
2842 for client in self._configuration_arg('player_client'):
2843 if client in allowed_clients:
2844 requested_clients.append(client)
2845 elif client == 'default':
2846 requested_clients.extend(default)
2847 elif client == 'all':
2848 requested_clients.extend(allowed_clients)
2849 else:
2850 self.report_warning(f'Skipping unsupported client {client}')
2851 if not requested_clients:
2852 requested_clients = default
2853
2854 if smuggled_data.get('is_music_url') or self.is_music_url(url):
2855 requested_clients.extend(
2856 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
2857
2858 return orderedSet(requested_clients)
2859
2860 def _extract_player_ytcfg(self, client, video_id):
2861 url = {
2862 'web_music': 'https://music.youtube.com',
2863 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
2864 }.get(client)
2865 if not url:
2866 return {}
2867 webpage = self._download_webpage(url, video_id, fatal=False, note='Downloading %s config' % client.replace('_', ' ').strip())
2868 return self.extract_ytcfg(video_id, webpage) or {}
2869
2870 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
2871 initial_pr = None
2872 if webpage:
2873 initial_pr = self._extract_yt_initial_variable(
2874 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
2875 video_id, 'initial player response')
2876
2877 original_clients = clients
2878 clients = clients[::-1]
2879 prs = []
2880
2881 def append_client(client_name):
2882 if client_name in INNERTUBE_CLIENTS and client_name not in original_clients:
2883 clients.append(client_name)
2884
2885 # Android player_response does not have microFormats which are needed for
2886 # extraction of some data. So we return the initial_pr with formats
2887 # stripped out even if not requested by the user
2888 # See: https://github.com/yt-dlp/yt-dlp/issues/501
2889 if initial_pr:
2890 pr = dict(initial_pr)
2891 pr['streamingData'] = None
2892 prs.append(pr)
2893
2894 last_error = None
2895 tried_iframe_fallback = False
2896 player_url = None
2897 while clients:
2898 client = clients.pop()
2899 player_ytcfg = master_ytcfg if client == 'web' else {}
2900 if 'configs' not in self._configuration_arg('player_skip'):
2901 player_ytcfg = self._extract_player_ytcfg(client, video_id) or player_ytcfg
2902
2903 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
2904 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
2905 if 'js' in self._configuration_arg('player_skip'):
2906 require_js_player = False
2907 player_url = None
2908
2909 if not player_url and not tried_iframe_fallback and require_js_player:
2910 player_url = self._download_player_url(video_id)
2911 tried_iframe_fallback = True
2912
2913 try:
2914 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
2915 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
2916 except ExtractorError as e:
2917 if last_error:
2918 self.report_warning(last_error)
2919 last_error = e
2920 continue
2921
2922 if pr:
2923 prs.append(pr)
2924
2925 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
2926 if client.endswith('_agegate') and self._is_unplayable(pr) and self.is_authenticated:
2927 append_client(client.replace('_agegate', '_creator'))
2928 elif self._is_agegated(pr):
2929 append_client(f'{client}_agegate')
2930
2931 if last_error:
2932 if not len(prs):
2933 raise last_error
2934 self.report_warning(last_error)
2935 return prs, player_url
2936
2937 def _extract_formats(self, streaming_data, video_id, player_url, is_live):
2938 itags, stream_ids = {}, []
2939 itag_qualities, res_qualities = {}, {}
2940 q = qualities([
2941 # Normally tiny is the smallest video-only formats. But
2942 # audio-only formats with unknown quality may get tagged as tiny
2943 'tiny',
2944 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
2945 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
2946 ])
2947 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
2948
2949 for fmt in streaming_formats:
2950 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
2951 continue
2952
2953 itag = str_or_none(fmt.get('itag'))
2954 audio_track = fmt.get('audioTrack') or {}
2955 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
2956 if stream_id in stream_ids:
2957 continue
2958
2959 quality = fmt.get('quality')
2960 height = int_or_none(fmt.get('height'))
2961 if quality == 'tiny' or not quality:
2962 quality = fmt.get('audioQuality', '').lower() or quality
2963 # The 3gp format (17) in android client has a quality of "small",
2964 # but is actually worse than other formats
2965 if itag == '17':
2966 quality = 'tiny'
2967 if quality:
2968 if itag:
2969 itag_qualities[itag] = quality
2970 if height:
2971 res_qualities[height] = quality
2972 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
2973 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
2974 # number of fragment that would subsequently requested with (`&sq=N`)
2975 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
2976 continue
2977
2978 fmt_url = fmt.get('url')
2979 if not fmt_url:
2980 sc = compat_parse_qs(fmt.get('signatureCipher'))
2981 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
2982 encrypted_sig = try_get(sc, lambda x: x['s'][0])
2983 if not (sc and fmt_url and encrypted_sig):
2984 continue
2985 if not player_url:
2986 continue
2987 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
2988 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
2989 fmt_url += '&' + sp + '=' + signature
2990
2991 query = parse_qs(fmt_url)
2992 throttled = False
2993 if query.get('n'):
2994 try:
2995 fmt_url = update_url_query(fmt_url, {
2996 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
2997 except ExtractorError as e:
2998 self.report_warning(
2999 f'nsig extraction failed: You may experience throttling for some formats\n'
3000 f'n = {query["n"][0]} ; player = {player_url}\n{e}', only_once=True)
3001 throttled = True
3002
3003 if itag:
3004 itags[itag] = 'https'
3005 stream_ids.append(stream_id)
3006
3007 tbr = float_or_none(
3008 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3009 language_preference = (
3010 10 if audio_track.get('audioIsDefault') and 10
3011 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3012 else -1)
3013 dct = {
3014 'asr': int_or_none(fmt.get('audioSampleRate')),
3015 'filesize': int_or_none(fmt.get('contentLength')),
3016 'format_id': itag,
3017 'format_note': join_nonempty(
3018 '%s%s' % (audio_track.get('displayName') or '',
3019 ' (default)' if language_preference > 0 else ''),
3020 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3021 throttled and 'THROTTLED', delim=', '),
3022 'source_preference': -10 if throttled else -1,
3023 'fps': int_or_none(fmt.get('fps')) or None,
3024 'height': height,
3025 'quality': q(quality),
3026 'tbr': tbr,
3027 'url': fmt_url,
3028 'width': int_or_none(fmt.get('width')),
3029 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3030 'desc' if language_preference < -1 else ''),
3031 'language_preference': language_preference,
3032 }
3033 mime_mobj = re.match(
3034 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3035 if mime_mobj:
3036 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3037 dct.update(parse_codecs(mime_mobj.group(2)))
3038 no_audio = dct.get('acodec') == 'none'
3039 no_video = dct.get('vcodec') == 'none'
3040 if no_audio:
3041 dct['vbr'] = tbr
3042 if no_video:
3043 dct['abr'] = tbr
3044 if no_audio or no_video:
3045 dct['downloader_options'] = {
3046 # Youtube throttles chunks >~10M
3047 'http_chunk_size': 10485760,
3048 }
3049 if dct.get('ext'):
3050 dct['container'] = dct['ext'] + '_dash'
3051 yield dct
3052
3053 live_from_start = is_live and self.get_param('live_from_start')
3054 skip_manifests = self._configuration_arg('skip')
3055 if not self.get_param('youtube_include_hls_manifest', True):
3056 skip_manifests.append('hls')
3057 get_dash = 'dash' not in skip_manifests and (
3058 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3059 get_hls = not live_from_start and 'hls' not in skip_manifests
3060
3061 def process_manifest_format(f, proto, itag):
3062 if itag in itags:
3063 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3064 return False
3065 itag = f'{itag}-{proto}'
3066 if itag:
3067 f['format_id'] = itag
3068 itags[itag] = proto
3069
3070 f['quality'] = next((
3071 q(qdict[val])
3072 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3073 if val in qdict), -1)
3074 return True
3075
3076 for sd in streaming_data:
3077 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3078 if hls_manifest_url:
3079 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3080 if process_manifest_format(f, 'hls', self._search_regex(
3081 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3082 yield f
3083
3084 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3085 if dash_manifest_url:
3086 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3087 if process_manifest_format(f, 'dash', f['format_id']):
3088 f['filesize'] = int_or_none(self._search_regex(
3089 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3090 if live_from_start:
3091 f['is_from_start'] = True
3092
3093 yield f
3094
3095 def _extract_storyboard(self, player_responses, duration):
3096 spec = get_first(
3097 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3098 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3099 if not base_url:
3100 return
3101 L = len(spec) - 1
3102 for i, args in enumerate(spec):
3103 args = args.split('#')
3104 counts = list(map(int_or_none, args[:5]))
3105 if len(args) != 8 or not all(counts):
3106 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3107 continue
3108 width, height, frame_count, cols, rows = counts
3109 N, sigh = args[6:]
3110
3111 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3112 fragment_count = frame_count / (cols * rows)
3113 fragment_duration = duration / fragment_count
3114 yield {
3115 'format_id': f'sb{i}',
3116 'format_note': 'storyboard',
3117 'ext': 'mhtml',
3118 'protocol': 'mhtml',
3119 'acodec': 'none',
3120 'vcodec': 'none',
3121 'url': url,
3122 'width': width,
3123 'height': height,
3124 'fragments': [{
3125 'path': url.replace('$M', str(j)),
3126 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3127 } for j in range(math.ceil(fragment_count))],
3128 }
3129
3130 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3131 webpage = None
3132 if 'webpage' not in self._configuration_arg('player_skip'):
3133 webpage = self._download_webpage(
3134 webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False)
3135
3136 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3137
3138 player_responses, player_url = self._extract_player_responses(
3139 self._get_requested_clients(url, smuggled_data),
3140 video_id, webpage, master_ytcfg)
3141
3142 return webpage, master_ytcfg, player_responses, player_url
3143
3144 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url):
3145 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3146 is_live = get_first(video_details, 'isLive')
3147 if is_live is None:
3148 is_live = get_first(live_broadcast_details, 'isLiveNow')
3149
3150 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3151 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live))
3152
3153 return live_broadcast_details, is_live, streaming_data, formats
3154
3155 def _real_extract(self, url):
3156 url, smuggled_data = unsmuggle_url(url, {})
3157 video_id = self._match_id(url)
3158
3159 base_url = self.http_scheme() + '//www.youtube.com/'
3160 webpage_url = base_url + 'watch?v=' + video_id
3161
3162 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3163
3164 playability_statuses = traverse_obj(
3165 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3166
3167 trailer_video_id = get_first(
3168 playability_statuses,
3169 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3170 expected_type=str)
3171 if trailer_video_id:
3172 return self.url_result(
3173 trailer_video_id, self.ie_key(), trailer_video_id)
3174
3175 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3176 if webpage else (lambda x: None))
3177
3178 video_details = traverse_obj(
3179 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3180 microformats = traverse_obj(
3181 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3182 expected_type=dict, default=[])
3183 video_title = (
3184 get_first(video_details, 'title')
3185 or self._get_text(microformats, (..., 'title'))
3186 or search_meta(['og:title', 'twitter:title', 'title']))
3187 video_description = get_first(video_details, 'shortDescription')
3188
3189 multifeed_metadata_list = get_first(
3190 player_responses,
3191 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3192 expected_type=str)
3193 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3194 if self.get_param('noplaylist'):
3195 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3196 else:
3197 entries = []
3198 feed_ids = []
3199 for feed in multifeed_metadata_list.split(','):
3200 # Unquote should take place before split on comma (,) since textual
3201 # fields may contain comma as well (see
3202 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3203 feed_data = compat_parse_qs(
3204 compat_urllib_parse_unquote_plus(feed))
3205
3206 def feed_entry(name):
3207 return try_get(
3208 feed_data, lambda x: x[name][0], compat_str)
3209
3210 feed_id = feed_entry('id')
3211 if not feed_id:
3212 continue
3213 feed_title = feed_entry('title')
3214 title = video_title
3215 if feed_title:
3216 title += ' (%s)' % feed_title
3217 entries.append({
3218 '_type': 'url_transparent',
3219 'ie_key': 'Youtube',
3220 'url': smuggle_url(
3221 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3222 {'force_singlefeed': True}),
3223 'title': title,
3224 })
3225 feed_ids.append(feed_id)
3226 self.to_screen(
3227 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3228 % (', '.join(feed_ids), video_id))
3229 return self.playlist_result(
3230 entries, video_id, video_title, video_description)
3231
3232 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(video_id, microformats, video_details, player_responses, player_url)
3233
3234 if not formats:
3235 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3236 self.report_drm(video_id)
3237 pemr = get_first(
3238 playability_statuses,
3239 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3240 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3241 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3242 if subreason:
3243 if subreason == 'The uploader has not made this video available in your country.':
3244 countries = get_first(microformats, 'availableCountries')
3245 if not countries:
3246 regions_allowed = search_meta('regionsAllowed')
3247 countries = regions_allowed.split(',') if regions_allowed else None
3248 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3249 reason += f'. {subreason}'
3250 if reason:
3251 self.raise_no_formats(reason, expected=True)
3252
3253 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3254 if not keywords and webpage:
3255 keywords = [
3256 unescapeHTML(m.group('content'))
3257 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3258 for keyword in keywords:
3259 if keyword.startswith('yt:stretch='):
3260 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3261 if mobj:
3262 # NB: float is intentional for forcing float division
3263 w, h = (float(v) for v in mobj.groups())
3264 if w > 0 and h > 0:
3265 ratio = w / h
3266 for f in formats:
3267 if f.get('vcodec') != 'none':
3268 f['stretched_ratio'] = ratio
3269 break
3270 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3271 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3272 if thumbnail_url:
3273 thumbnails.append({
3274 'url': thumbnail_url,
3275 })
3276 original_thumbnails = thumbnails.copy()
3277
3278 # The best resolution thumbnails sometimes does not appear in the webpage
3279 # See: https://github.com/ytdl-org/youtube-dl/issues/29049, https://github.com/yt-dlp/yt-dlp/issues/340
3280 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3281 thumbnail_names = [
3282 'maxresdefault', 'hq720', 'sddefault', 'sd1', 'sd2', 'sd3',
3283 'hqdefault', 'hq1', 'hq2', 'hq3', '0',
3284 'mqdefault', 'mq1', 'mq2', 'mq3',
3285 'default', '1', '2', '3'
3286 ]
3287 n_thumbnail_names = len(thumbnail_names)
3288 thumbnails.extend({
3289 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3290 video_id=video_id, name=name, ext=ext,
3291 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3292 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3293 for thumb in thumbnails:
3294 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3295 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3296 self._remove_duplicate_formats(thumbnails)
3297 self._downloader._sort_thumbnails(original_thumbnails)
3298
3299 category = get_first(microformats, 'category') or search_meta('genre')
3300 channel_id = str_or_none(
3301 get_first(video_details, 'channelId')
3302 or get_first(microformats, 'externalChannelId')
3303 or search_meta('channelId'))
3304 duration = int_or_none(
3305 get_first(video_details, 'lengthSeconds')
3306 or get_first(microformats, 'lengthSeconds')
3307 or parse_duration(search_meta('duration'))) or None
3308 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3309
3310 live_content = get_first(video_details, 'isLiveContent')
3311 is_upcoming = get_first(video_details, 'isUpcoming')
3312 if is_live is None:
3313 if is_upcoming or live_content is False:
3314 is_live = False
3315 if is_upcoming is None and (live_content or is_live):
3316 is_upcoming = False
3317 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3318 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3319 if not duration and live_end_time and live_start_time:
3320 duration = live_end_time - live_start_time
3321
3322 if is_live and self.get_param('live_from_start'):
3323 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3324
3325 formats.extend(self._extract_storyboard(player_responses, duration))
3326
3327 # Source is given priority since formats that throttle are given lower source_preference
3328 # When throttling issue is fully fixed, remove this
3329 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3330
3331 info = {
3332 'id': video_id,
3333 'title': video_title,
3334 'formats': formats,
3335 'thumbnails': thumbnails,
3336 # The best thumbnail that we are sure exists. Prevents unnecessary
3337 # URL checking if user don't care about getting the best possible thumbnail
3338 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3339 'description': video_description,
3340 'upload_date': unified_strdate(
3341 get_first(microformats, 'uploadDate')
3342 or search_meta('uploadDate')),
3343 'uploader': get_first(video_details, 'author'),
3344 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3345 'uploader_url': owner_profile_url,
3346 'channel_id': channel_id,
3347 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3348 'duration': duration,
3349 'view_count': int_or_none(
3350 get_first((video_details, microformats), (..., 'viewCount'))
3351 or search_meta('interactionCount')),
3352 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3353 'age_limit': 18 if (
3354 get_first(microformats, 'isFamilySafe') is False
3355 or search_meta('isFamilyFriendly') == 'false'
3356 or search_meta('og:restrictions:age') == '18+') else 0,
3357 'webpage_url': webpage_url,
3358 'categories': [category] if category else None,
3359 'tags': keywords,
3360 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3361 'is_live': is_live,
3362 'was_live': (False if is_live or is_upcoming or live_content is False
3363 else None if is_live is None or is_upcoming is None
3364 else live_content),
3365 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3366 'release_timestamp': live_start_time,
3367 }
3368
3369 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3370 if pctr:
3371 def get_lang_code(track):
3372 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3373 or track.get('languageCode'))
3374
3375 # Converted into dicts to remove duplicates
3376 captions = {
3377 get_lang_code(sub): sub
3378 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3379 translation_languages = {
3380 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3381 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3382
3383 def process_language(container, base_url, lang_code, sub_name, query):
3384 lang_subs = container.setdefault(lang_code, [])
3385 for fmt in self._SUBTITLE_FORMATS:
3386 query.update({
3387 'fmt': fmt,
3388 })
3389 lang_subs.append({
3390 'ext': fmt,
3391 'url': update_url_query(base_url, query),
3392 'name': sub_name,
3393 })
3394
3395 subtitles, automatic_captions = {}, {}
3396 for lang_code, caption_track in captions.items():
3397 base_url = caption_track.get('baseUrl')
3398 if not base_url:
3399 continue
3400 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3401 if caption_track.get('kind') != 'asr':
3402 if not lang_code:
3403 continue
3404 process_language(
3405 subtitles, base_url, lang_code, lang_name, {})
3406 if not caption_track.get('isTranslatable'):
3407 continue
3408 for trans_code, trans_name in translation_languages.items():
3409 if not trans_code:
3410 continue
3411 if caption_track.get('kind') != 'asr':
3412 trans_code += f'-{lang_code}'
3413 trans_name += format_field(lang_name, template=' from %s')
3414 process_language(
3415 automatic_captions, base_url, trans_code, trans_name, {'tlang': trans_code})
3416 if lang_code == f'a-{trans_code}':
3417 process_language(
3418 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {'tlang': trans_code})
3419 info['automatic_captions'] = automatic_captions
3420 info['subtitles'] = subtitles
3421
3422 parsed_url = compat_urllib_parse_urlparse(url)
3423 for component in [parsed_url.fragment, parsed_url.query]:
3424 query = compat_parse_qs(component)
3425 for k, v in query.items():
3426 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3427 d_k += '_time'
3428 if d_k not in info and k in s_ks:
3429 info[d_k] = parse_duration(query[k][0])
3430
3431 # Youtube Music Auto-generated description
3432 if video_description:
3433 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)
3434 if mobj:
3435 release_year = mobj.group('release_year')
3436 release_date = mobj.group('release_date')
3437 if release_date:
3438 release_date = release_date.replace('-', '')
3439 if not release_year:
3440 release_year = release_date[:4]
3441 info.update({
3442 'album': mobj.group('album'.strip()),
3443 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3444 'track': mobj.group('track').strip(),
3445 'release_date': release_date,
3446 'release_year': int_or_none(release_year),
3447 })
3448
3449 initial_data = None
3450 if webpage:
3451 initial_data = self._extract_yt_initial_variable(
3452 webpage, self._YT_INITIAL_DATA_RE, video_id,
3453 'yt initial data')
3454 if not initial_data:
3455 query = {'videoId': video_id}
3456 query.update(self._get_checkok_params())
3457 initial_data = self._extract_response(
3458 item_id=video_id, ep='next', fatal=False,
3459 ytcfg=master_ytcfg, query=query,
3460 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3461 note='Downloading initial data API JSON')
3462
3463 try:
3464 # This will error if there is no livechat
3465 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3466 info.setdefault('subtitles', {})['live_chat'] = [{
3467 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
3468 'video_id': video_id,
3469 'ext': 'json',
3470 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3471 }]
3472 except (KeyError, IndexError, TypeError):
3473 pass
3474
3475 if initial_data:
3476 info['chapters'] = (
3477 self._extract_chapters_from_json(initial_data, duration)
3478 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3479 or None)
3480
3481 contents = try_get(
3482 initial_data,
3483 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
3484 list) or []
3485 for content in contents:
3486 vpir = content.get('videoPrimaryInfoRenderer')
3487 if vpir:
3488 stl = vpir.get('superTitleLink')
3489 if stl:
3490 stl = self._get_text(stl)
3491 if try_get(
3492 vpir,
3493 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3494 info['location'] = stl
3495 else:
3496 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
3497 if mobj:
3498 info.update({
3499 'series': mobj.group(1),
3500 'season_number': int(mobj.group(2)),
3501 'episode_number': int(mobj.group(3)),
3502 })
3503 for tlb in (try_get(
3504 vpir,
3505 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3506 list) or []):
3507 tbr = tlb.get('toggleButtonRenderer') or {}
3508 for getter, regex in [(
3509 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3510 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3511 lambda x: x['accessibility'],
3512 lambda x: x['accessibilityData']['accessibilityData'],
3513 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3514 label = (try_get(tbr, getter, dict) or {}).get('label')
3515 if label:
3516 mobj = re.match(regex, label)
3517 if mobj:
3518 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3519 break
3520 sbr_tooltip = try_get(
3521 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3522 if sbr_tooltip:
3523 like_count, dislike_count = sbr_tooltip.split(' / ')
3524 info.update({
3525 'like_count': str_to_int(like_count),
3526 'dislike_count': str_to_int(dislike_count),
3527 })
3528 vsir = content.get('videoSecondaryInfoRenderer')
3529 if vsir:
3530 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3531 info.update({
3532 'channel': self._get_text(vor, 'title'),
3533 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3534
3535 rows = try_get(
3536 vsir,
3537 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3538 list) or []
3539 multiple_songs = False
3540 for row in rows:
3541 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3542 multiple_songs = True
3543 break
3544 for row in rows:
3545 mrr = row.get('metadataRowRenderer') or {}
3546 mrr_title = mrr.get('title')
3547 if not mrr_title:
3548 continue
3549 mrr_title = self._get_text(mrr, 'title')
3550 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3551 if mrr_title == 'License':
3552 info['license'] = mrr_contents_text
3553 elif not multiple_songs:
3554 if mrr_title == 'Album':
3555 info['album'] = mrr_contents_text
3556 elif mrr_title == 'Artist':
3557 info['artist'] = mrr_contents_text
3558 elif mrr_title == 'Song':
3559 info['track'] = mrr_contents_text
3560
3561 fallbacks = {
3562 'channel': 'uploader',
3563 'channel_id': 'uploader_id',
3564 'channel_url': 'uploader_url',
3565 }
3566 for to, frm in fallbacks.items():
3567 if not info.get(to):
3568 info[to] = info.get(frm)
3569
3570 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3571 v = info.get(s_k)
3572 if v:
3573 info[d_k] = v
3574
3575 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3576 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3577 is_membersonly = None
3578 is_premium = None
3579 if initial_data and is_private is not None:
3580 is_membersonly = False
3581 is_premium = False
3582 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3583 badge_labels = set()
3584 for content in contents:
3585 if not isinstance(content, dict):
3586 continue
3587 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3588 for badge_label in badge_labels:
3589 if badge_label.lower() == 'members only':
3590 is_membersonly = True
3591 elif badge_label.lower() == 'premium':
3592 is_premium = True
3593 elif badge_label.lower() == 'unlisted':
3594 is_unlisted = True
3595
3596 info['availability'] = self._availability(
3597 is_private=is_private,
3598 needs_premium=is_premium,
3599 needs_subscription=is_membersonly,
3600 needs_auth=info['age_limit'] >= 18,
3601 is_unlisted=None if is_private is None else is_unlisted)
3602
3603 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3604
3605 self.mark_watched(video_id, player_responses)
3606
3607 return info
3608
3609
3610 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3611
3612 @staticmethod
3613 def passthrough_smuggled_data(func):
3614 def _smuggle(entries, smuggled_data):
3615 for entry in entries:
3616 # TODO: Convert URL to music.youtube instead.
3617 # Do we need to passthrough any other smuggled_data?
3618 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3619 yield entry
3620
3621 @functools.wraps(func)
3622 def wrapper(self, url):
3623 url, smuggled_data = unsmuggle_url(url, {})
3624 if self.is_music_url(url):
3625 smuggled_data['is_music_url'] = True
3626 info_dict = func(self, url, smuggled_data)
3627 if smuggled_data and info_dict.get('entries'):
3628 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3629 return info_dict
3630 return wrapper
3631
3632 def _extract_channel_id(self, webpage):
3633 channel_id = self._html_search_meta(
3634 'channelId', webpage, 'channel id', default=None)
3635 if channel_id:
3636 return channel_id
3637 channel_url = self._html_search_meta(
3638 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3639 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3640 'twitter:app:url:googleplay'), webpage, 'channel url')
3641 return self._search_regex(
3642 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3643 channel_url, 'channel id')
3644
3645 @staticmethod
3646 def _extract_basic_item_renderer(item):
3647 # Modified from _extract_grid_item_renderer
3648 known_basic_renderers = (
3649 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer'
3650 )
3651 for key, renderer in item.items():
3652 if not isinstance(renderer, dict):
3653 continue
3654 elif key in known_basic_renderers:
3655 return renderer
3656 elif key.startswith('grid') and key.endswith('Renderer'):
3657 return renderer
3658
3659 def _grid_entries(self, grid_renderer):
3660 for item in grid_renderer['items']:
3661 if not isinstance(item, dict):
3662 continue
3663 renderer = self._extract_basic_item_renderer(item)
3664 if not isinstance(renderer, dict):
3665 continue
3666 title = self._get_text(renderer, 'title')
3667
3668 # playlist
3669 playlist_id = renderer.get('playlistId')
3670 if playlist_id:
3671 yield self.url_result(
3672 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3673 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3674 video_title=title)
3675 continue
3676 # video
3677 video_id = renderer.get('videoId')
3678 if video_id:
3679 yield self._extract_video(renderer)
3680 continue
3681 # channel
3682 channel_id = renderer.get('channelId')
3683 if channel_id:
3684 yield self.url_result(
3685 'https://www.youtube.com/channel/%s' % channel_id,
3686 ie=YoutubeTabIE.ie_key(), video_title=title)
3687 continue
3688 # generic endpoint URL support
3689 ep_url = urljoin('https://www.youtube.com/', try_get(
3690 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3691 compat_str))
3692 if ep_url:
3693 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3694 if ie.suitable(ep_url):
3695 yield self.url_result(
3696 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3697 break
3698
3699 def _music_reponsive_list_entry(self, renderer):
3700 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3701 if video_id:
3702 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3703 ie=YoutubeIE.ie_key(), video_id=video_id)
3704 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3705 if playlist_id:
3706 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3707 if video_id:
3708 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3709 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3710 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3711 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3712 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3713 if browse_id:
3714 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3715 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3716
3717 def _shelf_entries_from_content(self, shelf_renderer):
3718 content = shelf_renderer.get('content')
3719 if not isinstance(content, dict):
3720 return
3721 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3722 if renderer:
3723 # TODO: add support for nested playlists so each shelf is processed
3724 # as separate playlist
3725 # TODO: this includes only first N items
3726 for entry in self._grid_entries(renderer):
3727 yield entry
3728 renderer = content.get('horizontalListRenderer')
3729 if renderer:
3730 # TODO
3731 pass
3732
3733 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3734 ep = try_get(
3735 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3736 compat_str)
3737 shelf_url = urljoin('https://www.youtube.com', ep)
3738 if shelf_url:
3739 # Skipping links to another channels, note that checking for
3740 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3741 # will not work
3742 if skip_channels and '/channels?' in shelf_url:
3743 return
3744 title = self._get_text(shelf_renderer, 'title')
3745 yield self.url_result(shelf_url, video_title=title)
3746 # Shelf may not contain shelf URL, fallback to extraction from content
3747 for entry in self._shelf_entries_from_content(shelf_renderer):
3748 yield entry
3749
3750 def _playlist_entries(self, video_list_renderer):
3751 for content in video_list_renderer['contents']:
3752 if not isinstance(content, dict):
3753 continue
3754 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3755 if not isinstance(renderer, dict):
3756 continue
3757 video_id = renderer.get('videoId')
3758 if not video_id:
3759 continue
3760 yield self._extract_video(renderer)
3761
3762 def _rich_entries(self, rich_grid_renderer):
3763 renderer = try_get(
3764 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3765 video_id = renderer.get('videoId')
3766 if not video_id:
3767 return
3768 yield self._extract_video(renderer)
3769
3770 def _video_entry(self, video_renderer):
3771 video_id = video_renderer.get('videoId')
3772 if video_id:
3773 return self._extract_video(video_renderer)
3774
3775 def _post_thread_entries(self, post_thread_renderer):
3776 post_renderer = try_get(
3777 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3778 if not post_renderer:
3779 return
3780 # video attachment
3781 video_renderer = try_get(
3782 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3783 video_id = video_renderer.get('videoId')
3784 if video_id:
3785 entry = self._extract_video(video_renderer)
3786 if entry:
3787 yield entry
3788 # playlist attachment
3789 playlist_id = try_get(
3790 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3791 if playlist_id:
3792 yield self.url_result(
3793 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3794 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3795 # inline video links
3796 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
3797 for run in runs:
3798 if not isinstance(run, dict):
3799 continue
3800 ep_url = try_get(
3801 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
3802 if not ep_url:
3803 continue
3804 if not YoutubeIE.suitable(ep_url):
3805 continue
3806 ep_video_id = YoutubeIE._match_id(ep_url)
3807 if video_id == ep_video_id:
3808 continue
3809 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
3810
3811 def _post_thread_continuation_entries(self, post_thread_continuation):
3812 contents = post_thread_continuation.get('contents')
3813 if not isinstance(contents, list):
3814 return
3815 for content in contents:
3816 renderer = content.get('backstagePostThreadRenderer')
3817 if not isinstance(renderer, dict):
3818 continue
3819 for entry in self._post_thread_entries(renderer):
3820 yield entry
3821
3822 r''' # unused
3823 def _rich_grid_entries(self, contents):
3824 for content in contents:
3825 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
3826 if video_renderer:
3827 entry = self._video_entry(video_renderer)
3828 if entry:
3829 yield entry
3830 '''
3831 def _extract_entries(self, parent_renderer, continuation_list):
3832 # continuation_list is modified in-place with continuation_list = [continuation_token]
3833 continuation_list[:] = [None]
3834 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
3835 for content in contents:
3836 if not isinstance(content, dict):
3837 continue
3838 is_renderer = traverse_obj(
3839 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
3840 expected_type=dict)
3841 if not is_renderer:
3842 renderer = content.get('richItemRenderer')
3843 if renderer:
3844 for entry in self._rich_entries(renderer):
3845 yield entry
3846 continuation_list[0] = self._extract_continuation(parent_renderer)
3847 continue
3848 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
3849 for isr_content in isr_contents:
3850 if not isinstance(isr_content, dict):
3851 continue
3852
3853 known_renderers = {
3854 'playlistVideoListRenderer': self._playlist_entries,
3855 'gridRenderer': self._grid_entries,
3856 'shelfRenderer': lambda x: self._shelf_entries(x),
3857 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
3858 'backstagePostThreadRenderer': self._post_thread_entries,
3859 'videoRenderer': lambda x: [self._video_entry(x)],
3860 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
3861 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
3862 }
3863 for key, renderer in isr_content.items():
3864 if key not in known_renderers:
3865 continue
3866 for entry in known_renderers[key](renderer):
3867 if entry:
3868 yield entry
3869 continuation_list[0] = self._extract_continuation(renderer)
3870 break
3871
3872 if not continuation_list[0]:
3873 continuation_list[0] = self._extract_continuation(is_renderer)
3874
3875 if not continuation_list[0]:
3876 continuation_list[0] = self._extract_continuation(parent_renderer)
3877
3878 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
3879 continuation_list = [None]
3880 extract_entries = lambda x: self._extract_entries(x, continuation_list)
3881 tab_content = try_get(tab, lambda x: x['content'], dict)
3882 if not tab_content:
3883 return
3884 parent_renderer = (
3885 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
3886 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
3887 for entry in extract_entries(parent_renderer):
3888 yield entry
3889 continuation = continuation_list[0]
3890
3891 for page_num in itertools.count(1):
3892 if not continuation:
3893 break
3894 headers = self.generate_api_headers(
3895 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
3896 response = self._extract_response(
3897 item_id='%s page %s' % (item_id, page_num),
3898 query=continuation, headers=headers, ytcfg=ytcfg,
3899 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3900
3901 if not response:
3902 break
3903 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
3904 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
3905 visitor_data = self._extract_visitor_data(response) or visitor_data
3906
3907 known_continuation_renderers = {
3908 'playlistVideoListContinuation': self._playlist_entries,
3909 'gridContinuation': self._grid_entries,
3910 'itemSectionContinuation': self._post_thread_continuation_entries,
3911 'sectionListContinuation': extract_entries, # for feeds
3912 }
3913 continuation_contents = try_get(
3914 response, lambda x: x['continuationContents'], dict) or {}
3915 continuation_renderer = None
3916 for key, value in continuation_contents.items():
3917 if key not in known_continuation_renderers:
3918 continue
3919 continuation_renderer = value
3920 continuation_list = [None]
3921 for entry in known_continuation_renderers[key](continuation_renderer):
3922 yield entry
3923 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
3924 break
3925 if continuation_renderer:
3926 continue
3927
3928 known_renderers = {
3929 'gridPlaylistRenderer': (self._grid_entries, 'items'),
3930 'gridVideoRenderer': (self._grid_entries, 'items'),
3931 'gridChannelRenderer': (self._grid_entries, 'items'),
3932 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
3933 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
3934 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
3935 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
3936 }
3937 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3938 continuation_items = try_get(
3939 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
3940 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
3941 video_items_renderer = None
3942 for key, value in continuation_item.items():
3943 if key not in known_renderers:
3944 continue
3945 video_items_renderer = {known_renderers[key][1]: continuation_items}
3946 continuation_list = [None]
3947 for entry in known_renderers[key][0](video_items_renderer):
3948 yield entry
3949 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3950 break
3951 if video_items_renderer:
3952 continue
3953 break
3954
3955 @staticmethod
3956 def _extract_selected_tab(tabs):
3957 for tab in tabs:
3958 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
3959 if renderer.get('selected') is True:
3960 return renderer
3961 else:
3962 raise ExtractorError('Unable to find selected tab')
3963
3964 @classmethod
3965 def _extract_uploader(cls, data):
3966 uploader = {}
3967 renderer = cls._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
3968 owner = try_get(
3969 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3970 if owner:
3971 uploader['uploader'] = owner.get('text')
3972 uploader['uploader_id'] = try_get(
3973 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3974 uploader['uploader_url'] = urljoin(
3975 'https://www.youtube.com/',
3976 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3977 return {k: v for k, v in uploader.items() if v is not None}
3978
3979 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
3980 playlist_id = title = description = channel_url = channel_name = channel_id = None
3981 tags = []
3982
3983 selected_tab = self._extract_selected_tab(tabs)
3984 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
3985 renderer = try_get(
3986 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3987 if renderer:
3988 channel_name = renderer.get('title')
3989 channel_url = renderer.get('channelUrl')
3990 channel_id = renderer.get('externalId')
3991 else:
3992 renderer = try_get(
3993 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3994
3995 if renderer:
3996 title = renderer.get('title')
3997 description = renderer.get('description', '')
3998 playlist_id = channel_id
3999 tags = renderer.get('keywords', '').split()
4000
4001 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4002 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4003 def _get_uncropped(url):
4004 return url_or_none((url or '').split('=')[0] + '=s0')
4005
4006 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4007 if avatar_thumbnails:
4008 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4009 if uncropped_avatar:
4010 avatar_thumbnails.append({
4011 'url': uncropped_avatar,
4012 'id': 'avatar_uncropped',
4013 'preference': 1
4014 })
4015
4016 channel_banners = self._extract_thumbnails(
4017 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4018 for banner in channel_banners:
4019 banner['preference'] = -10
4020
4021 if channel_banners:
4022 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4023 if uncropped_banner:
4024 channel_banners.append({
4025 'url': uncropped_banner,
4026 'id': 'banner_uncropped',
4027 'preference': -5
4028 })
4029
4030 primary_thumbnails = self._extract_thumbnails(
4031 primary_sidebar_renderer, ('thumbnailRenderer', 'playlistVideoThumbnailRenderer', 'thumbnail'))
4032
4033 if playlist_id is None:
4034 playlist_id = item_id
4035
4036 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4037 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4038 if title is None:
4039 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4040 title += format_field(selected_tab, 'title', ' - %s')
4041 title += format_field(selected_tab, 'expandedText', ' - %s')
4042
4043 metadata = {
4044 'playlist_id': playlist_id,
4045 'playlist_title': title,
4046 'playlist_description': description,
4047 'uploader': channel_name,
4048 'uploader_id': channel_id,
4049 'uploader_url': channel_url,
4050 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4051 'tags': tags,
4052 'view_count': self._get_count(playlist_stats, 1),
4053 'availability': self._extract_availability(data),
4054 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4055 'playlist_count': self._get_count(playlist_stats, 0),
4056 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4057 }
4058 if not channel_id:
4059 metadata.update(self._extract_uploader(data))
4060 metadata.update({
4061 'channel': metadata['uploader'],
4062 'channel_id': metadata['uploader_id'],
4063 'channel_url': metadata['uploader_url']})
4064 return self.playlist_result(
4065 self._entries(
4066 selected_tab, playlist_id, ytcfg,
4067 self._extract_account_syncid(ytcfg, data),
4068 self._extract_visitor_data(data, ytcfg)),
4069 **metadata)
4070
4071 def _extract_mix_playlist(self, playlist, playlist_id, data, ytcfg):
4072 first_id = last_id = response = None
4073 for page_num in itertools.count(1):
4074 videos = list(self._playlist_entries(playlist))
4075 if not videos:
4076 return
4077 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4078 if start >= len(videos):
4079 return
4080 for video in videos[start:]:
4081 if video['id'] == first_id:
4082 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
4083 return
4084 yield video
4085 first_id = first_id or videos[0]['id']
4086 last_id = videos[-1]['id']
4087 watch_endpoint = try_get(
4088 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4089 headers = self.generate_api_headers(
4090 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4091 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4092 query = {
4093 'playlistId': playlist_id,
4094 'videoId': watch_endpoint.get('videoId') or last_id,
4095 'index': watch_endpoint.get('index') or len(videos),
4096 'params': watch_endpoint.get('params') or 'OAE%3D'
4097 }
4098 response = self._extract_response(
4099 item_id='%s page %d' % (playlist_id, page_num),
4100 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4101 check_get_keys='contents'
4102 )
4103 playlist = try_get(
4104 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4105
4106 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4107 title = playlist.get('title') or try_get(
4108 data, lambda x: x['titleText']['simpleText'], compat_str)
4109 playlist_id = playlist.get('playlistId') or item_id
4110
4111 # Delegating everything except mix playlists to regular tab-based playlist URL
4112 playlist_url = urljoin(url, try_get(
4113 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4114 compat_str))
4115 if playlist_url and playlist_url != url:
4116 return self.url_result(
4117 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4118 video_title=title)
4119
4120 return self.playlist_result(
4121 self._extract_mix_playlist(playlist, playlist_id, data, ytcfg),
4122 playlist_id=playlist_id, playlist_title=title)
4123
4124 def _extract_availability(self, data):
4125 """
4126 Gets the availability of a given playlist/tab.
4127 Note: Unless YouTube tells us explicitly, we do not assume it is public
4128 @param data: response
4129 """
4130 is_private = is_unlisted = None
4131 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4132 badge_labels = self._extract_badges(renderer)
4133
4134 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4135 privacy_dropdown_entries = try_get(
4136 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4137 for renderer_dict in privacy_dropdown_entries:
4138 is_selected = try_get(
4139 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4140 if not is_selected:
4141 continue
4142 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4143 if label:
4144 badge_labels.add(label.lower())
4145 break
4146
4147 for badge_label in badge_labels:
4148 if badge_label == 'unlisted':
4149 is_unlisted = True
4150 elif badge_label == 'private':
4151 is_private = True
4152 elif badge_label == 'public':
4153 is_unlisted = is_private = False
4154 return self._availability(is_private, False, False, False, is_unlisted)
4155
4156 @staticmethod
4157 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4158 sidebar_renderer = try_get(
4159 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4160 for item in sidebar_renderer:
4161 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4162 if renderer:
4163 return renderer
4164
4165 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4166 """
4167 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4168 """
4169 browse_id = params = None
4170 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4171 if not renderer:
4172 return
4173 menu_renderer = try_get(
4174 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4175 for menu_item in menu_renderer:
4176 if not isinstance(menu_item, dict):
4177 continue
4178 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4179 text = try_get(
4180 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4181 if not text or text.lower() != 'show unavailable videos':
4182 continue
4183 browse_endpoint = try_get(
4184 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4185 browse_id = browse_endpoint.get('browseId')
4186 params = browse_endpoint.get('params')
4187 break
4188
4189 headers = self.generate_api_headers(
4190 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4191 visitor_data=self._extract_visitor_data(data, ytcfg))
4192 query = {
4193 'params': params or 'wgYCCAA=',
4194 'browseId': browse_id or 'VL%s' % item_id
4195 }
4196 return self._extract_response(
4197 item_id=item_id, headers=headers, query=query,
4198 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4199 note='Downloading API JSON with unavailable videos')
4200
4201 def _extract_webpage(self, url, item_id, fatal=True):
4202 retries = self.get_param('extractor_retries', 3)
4203 count = -1
4204 webpage = data = last_error = None
4205 while count < retries:
4206 count += 1
4207 # Sometimes youtube returns a webpage with incomplete ytInitialData
4208 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4209 if last_error:
4210 self.report_warning('%s. Retrying ...' % last_error)
4211 try:
4212 webpage = self._download_webpage(
4213 url, item_id,
4214 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4215 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4216 except ExtractorError as e:
4217 if isinstance(e.cause, network_exceptions):
4218 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4219 last_error = error_to_compat_str(e.cause or e.msg)
4220 if count < retries:
4221 continue
4222 if fatal:
4223 raise
4224 self.report_warning(error_to_compat_str(e))
4225 break
4226 else:
4227 try:
4228 self._extract_and_report_alerts(data)
4229 except ExtractorError as e:
4230 if fatal:
4231 raise
4232 self.report_warning(error_to_compat_str(e))
4233 break
4234
4235 if dict_get(data, ('contents', 'currentVideoEndpoint')):
4236 break
4237
4238 last_error = 'Incomplete yt initial data received'
4239 if count >= retries:
4240 if fatal:
4241 raise ExtractorError(last_error)
4242 self.report_warning(last_error)
4243 break
4244
4245 return webpage, data
4246
4247 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4248 data = None
4249 if 'webpage' not in self._configuration_arg('skip'):
4250 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4251 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4252 if not data:
4253 if not ytcfg and self.is_authenticated:
4254 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.'
4255 if 'authcheck' not in self._configuration_arg('skip') and fatal:
4256 raise ExtractorError(
4257 msg + ' If you are not downloading private content, or your cookies are only for the first account and channel,'
4258 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4259 expected=True)
4260 self.report_warning(msg, only_once=True)
4261 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4262 return data, ytcfg
4263
4264 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4265 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4266 resolve_response = self._extract_response(
4267 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4268 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4269 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4270 for ep_key, ep in endpoints.items():
4271 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4272 if params:
4273 return self._extract_response(
4274 item_id=item_id, query=params, ep=ep, headers=headers,
4275 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4276 check_get_keys=('contents', 'currentVideoEndpoint'))
4277 err_note = 'Failed to resolve url (does the playlist exist?)'
4278 if fatal:
4279 raise ExtractorError(err_note, expected=True)
4280 self.report_warning(err_note, item_id)
4281
4282 _SEARCH_PARAMS = None
4283
4284 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4285 data = {'query': query}
4286 if params is NO_DEFAULT:
4287 params = self._SEARCH_PARAMS
4288 if params:
4289 data['params'] = params
4290
4291 content_keys = (
4292 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4293 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4294 # ytmusic search
4295 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4296 ('continuationContents', ),
4297 )
4298 check_get_keys = tuple(set(keys[0] for keys in content_keys))
4299
4300 continuation_list = [None]
4301 for page_num in itertools.count(1):
4302 data.update(continuation_list[0] or {})
4303 search = self._extract_response(
4304 item_id='query "%s" page %s' % (query, page_num), ep='search', query=data,
4305 default_client=default_client, check_get_keys=check_get_keys)
4306 slr_contents = traverse_obj(search, *content_keys)
4307 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4308 if not continuation_list[0]:
4309 break
4310
4311
4312 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4313 IE_DESC = 'YouTube Tabs'
4314 _VALID_URL = r'''(?x:
4315 https?://
4316 (?:\w+\.)?
4317 (?:
4318 youtube(?:kids)?\.com|
4319 %(invidious)s
4320 )/
4321 (?:
4322 (?P<channel_type>channel|c|user|browse)/|
4323 (?P<not_channel>
4324 feed/|hashtag/|
4325 (?:playlist|watch)\?.*?\blist=
4326 )|
4327 (?!(?:%(reserved_names)s)\b) # Direct URLs
4328 )
4329 (?P<id>[^/?\#&]+)
4330 )''' % {
4331 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4332 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4333 }
4334 IE_NAME = 'youtube:tab'
4335
4336 _TESTS = [{
4337 'note': 'playlists, multipage',
4338 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4339 'playlist_mincount': 94,
4340 'info_dict': {
4341 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4342 'title': 'Igor Kleiner - Playlists',
4343 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4344 'uploader': 'Igor Kleiner',
4345 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4346 'channel': 'Igor Kleiner',
4347 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4348 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4349 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4350 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4351 'channel_follower_count': int
4352 },
4353 }, {
4354 'note': 'playlists, multipage, different order',
4355 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4356 'playlist_mincount': 94,
4357 'info_dict': {
4358 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4359 'title': 'Igor Kleiner - Playlists',
4360 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4361 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4362 'uploader': 'Igor Kleiner',
4363 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4364 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4365 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4366 'channel': 'Igor Kleiner',
4367 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4368 'channel_follower_count': int
4369 },
4370 }, {
4371 'note': 'playlists, series',
4372 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4373 'playlist_mincount': 5,
4374 'info_dict': {
4375 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4376 'title': '3Blue1Brown - Playlists',
4377 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4378 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4379 'uploader': '3Blue1Brown',
4380 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4381 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4382 'channel': '3Blue1Brown',
4383 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4384 'tags': ['Mathematics'],
4385 'channel_follower_count': int
4386 },
4387 }, {
4388 'note': 'playlists, singlepage',
4389 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4390 'playlist_mincount': 4,
4391 'info_dict': {
4392 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4393 'title': 'ThirstForScience - Playlists',
4394 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4395 'uploader': 'ThirstForScience',
4396 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4397 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4398 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4399 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4400 'tags': 'count:13',
4401 'channel': 'ThirstForScience',
4402 'channel_follower_count': int
4403 }
4404 }, {
4405 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4406 'only_matching': True,
4407 }, {
4408 'note': 'basic, single video playlist',
4409 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4410 'info_dict': {
4411 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4412 'uploader': 'Sergey M.',
4413 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4414 'title': 'youtube-dl public playlist',
4415 'description': '',
4416 'tags': [],
4417 'view_count': int,
4418 'modified_date': '20201130',
4419 'channel': 'Sergey M.',
4420 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4421 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4422 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4423 },
4424 'playlist_count': 1,
4425 }, {
4426 'note': 'empty playlist',
4427 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4428 'info_dict': {
4429 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4430 'uploader': 'Sergey M.',
4431 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4432 'title': 'youtube-dl empty playlist',
4433 'tags': [],
4434 'channel': 'Sergey M.',
4435 'description': '',
4436 'modified_date': '20160902',
4437 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4438 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4439 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4440 },
4441 'playlist_count': 0,
4442 }, {
4443 'note': 'Home tab',
4444 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4445 'info_dict': {
4446 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4447 'title': 'lex will - Home',
4448 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4449 'uploader': 'lex will',
4450 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4451 'channel': 'lex will',
4452 'tags': ['bible', 'history', 'prophesy'],
4453 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4454 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4455 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4456 'channel_follower_count': int
4457 },
4458 'playlist_mincount': 2,
4459 }, {
4460 'note': 'Videos tab',
4461 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4462 'info_dict': {
4463 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4464 'title': 'lex will - Videos',
4465 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4466 'uploader': 'lex will',
4467 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4468 'tags': ['bible', 'history', 'prophesy'],
4469 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4470 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4471 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4472 'channel': 'lex will',
4473 'channel_follower_count': int
4474 },
4475 'playlist_mincount': 975,
4476 }, {
4477 'note': 'Videos tab, sorted by popular',
4478 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4479 'info_dict': {
4480 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4481 'title': 'lex will - Videos',
4482 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4483 'uploader': 'lex will',
4484 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4485 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4486 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4487 'channel': 'lex will',
4488 'tags': ['bible', 'history', 'prophesy'],
4489 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4490 'channel_follower_count': int
4491 },
4492 'playlist_mincount': 199,
4493 }, {
4494 'note': 'Playlists tab',
4495 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4496 'info_dict': {
4497 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4498 'title': 'lex will - Playlists',
4499 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4500 'uploader': 'lex will',
4501 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4502 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4503 'channel': 'lex will',
4504 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4505 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4506 'tags': ['bible', 'history', 'prophesy'],
4507 'channel_follower_count': int
4508 },
4509 'playlist_mincount': 17,
4510 }, {
4511 'note': 'Community tab',
4512 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4513 'info_dict': {
4514 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4515 'title': 'lex will - Community',
4516 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4517 'uploader': 'lex will',
4518 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4519 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4520 'channel': 'lex will',
4521 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4522 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4523 'tags': ['bible', 'history', 'prophesy'],
4524 'channel_follower_count': int
4525 },
4526 'playlist_mincount': 18,
4527 }, {
4528 'note': 'Channels tab',
4529 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4530 'info_dict': {
4531 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4532 'title': 'lex will - Channels',
4533 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4534 'uploader': 'lex will',
4535 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4536 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4537 'channel': 'lex will',
4538 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4539 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4540 'tags': ['bible', 'history', 'prophesy'],
4541 'channel_follower_count': int
4542 },
4543 'playlist_mincount': 12,
4544 }, {
4545 'note': 'Search tab',
4546 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4547 'playlist_mincount': 40,
4548 'info_dict': {
4549 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4550 'title': '3Blue1Brown - Search - linear algebra',
4551 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4552 'uploader': '3Blue1Brown',
4553 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4554 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4555 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4556 'tags': ['Mathematics'],
4557 'channel': '3Blue1Brown',
4558 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4559 'channel_follower_count': int
4560 },
4561 }, {
4562 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4563 'only_matching': True,
4564 }, {
4565 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4566 'only_matching': True,
4567 }, {
4568 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4569 'only_matching': True,
4570 }, {
4571 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4572 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4573 'info_dict': {
4574 'title': '29C3: Not my department',
4575 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4576 'uploader': 'Christiaan008',
4577 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4578 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4579 'tags': [],
4580 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4581 'view_count': int,
4582 'modified_date': '20150605',
4583 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4584 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4585 'channel': 'Christiaan008',
4586 },
4587 'playlist_count': 96,
4588 }, {
4589 'note': 'Large playlist',
4590 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4591 'info_dict': {
4592 'title': 'Uploads from Cauchemar',
4593 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4594 'uploader': 'Cauchemar',
4595 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4596 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4597 'tags': [],
4598 'modified_date': r're:\d{8}',
4599 'channel': 'Cauchemar',
4600 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4601 'view_count': int,
4602 'description': '',
4603 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4604 },
4605 'playlist_mincount': 1123,
4606 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4607 }, {
4608 'note': 'even larger playlist, 8832 videos',
4609 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4610 'only_matching': True,
4611 }, {
4612 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4613 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4614 'info_dict': {
4615 'title': 'Uploads from Interstellar Movie',
4616 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4617 'uploader': 'Interstellar Movie',
4618 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4619 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4620 'tags': [],
4621 'view_count': int,
4622 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4623 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4624 'channel': 'Interstellar Movie',
4625 'description': '',
4626 'modified_date': r're:\d{8}',
4627 },
4628 'playlist_mincount': 21,
4629 }, {
4630 'note': 'Playlist with "show unavailable videos" button',
4631 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4632 'info_dict': {
4633 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4634 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4635 'uploader': 'Phim Siêu Nhân Nhật Bản',
4636 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4637 'view_count': int,
4638 'channel': 'Phim Siêu Nhân Nhật Bản',
4639 'tags': [],
4640 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4641 'description': '',
4642 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4643 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4644 'modified_date': r're:\d{8}',
4645 },
4646 'playlist_mincount': 200,
4647 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4648 }, {
4649 'note': 'Playlist with unavailable videos in page 7',
4650 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4651 'info_dict': {
4652 'title': 'Uploads from BlankTV',
4653 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4654 'uploader': 'BlankTV',
4655 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4656 'channel': 'BlankTV',
4657 'channel_url': 'https://www.youtube.com/c/blanktv',
4658 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4659 'view_count': int,
4660 'tags': [],
4661 'uploader_url': 'https://www.youtube.com/c/blanktv',
4662 'modified_date': r're:\d{8}',
4663 'description': '',
4664 },
4665 'playlist_mincount': 1000,
4666 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4667 }, {
4668 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4669 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4670 'info_dict': {
4671 'title': 'Data Analysis with Dr Mike Pound',
4672 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4673 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4674 'uploader': 'Computerphile',
4675 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4676 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4677 'tags': [],
4678 'view_count': int,
4679 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4680 'channel_url': 'https://www.youtube.com/user/Computerphile',
4681 'channel': 'Computerphile',
4682 },
4683 'playlist_mincount': 11,
4684 }, {
4685 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4686 'only_matching': True,
4687 }, {
4688 'note': 'Playlist URL that does not actually serve a playlist',
4689 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4690 'info_dict': {
4691 'id': 'FqZTN594JQw',
4692 'ext': 'webm',
4693 'title': "Smiley's People 01 detective, Adventure Series, Action",
4694 'uploader': 'STREEM',
4695 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4696 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4697 'upload_date': '20150526',
4698 'license': 'Standard YouTube License',
4699 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4700 'categories': ['People & Blogs'],
4701 'tags': list,
4702 'view_count': int,
4703 'like_count': int,
4704 },
4705 'params': {
4706 'skip_download': True,
4707 },
4708 'skip': 'This video is not available.',
4709 'add_ie': [YoutubeIE.ie_key()],
4710 }, {
4711 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4712 'only_matching': True,
4713 }, {
4714 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4715 'only_matching': True,
4716 }, {
4717 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4718 'info_dict': {
4719 'id': 'GgL890LIznQ', # This will keep changing
4720 'ext': 'mp4',
4721 'title': str,
4722 'uploader': 'Sky News',
4723 'uploader_id': 'skynews',
4724 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4725 'upload_date': r're:\d{8}',
4726 'description': str,
4727 'categories': ['News & Politics'],
4728 'tags': list,
4729 'like_count': int,
4730 'release_timestamp': 1642502819,
4731 'channel': 'Sky News',
4732 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4733 'age_limit': 0,
4734 'view_count': int,
4735 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4736 'playable_in_embed': True,
4737 'release_date': '20220118',
4738 'availability': 'public',
4739 'live_status': 'is_live',
4740 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4741 'channel_follower_count': int
4742 },
4743 'params': {
4744 'skip_download': True,
4745 },
4746 'expected_warnings': ['Ignoring subtitle tracks found in '],
4747 }, {
4748 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4749 'info_dict': {
4750 'id': 'a48o2S1cPoo',
4751 'ext': 'mp4',
4752 'title': 'The Young Turks - Live Main Show',
4753 'uploader': 'The Young Turks',
4754 'uploader_id': 'TheYoungTurks',
4755 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4756 'upload_date': '20150715',
4757 'license': 'Standard YouTube License',
4758 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4759 'categories': ['News & Politics'],
4760 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4761 'like_count': int,
4762 },
4763 'params': {
4764 'skip_download': True,
4765 },
4766 'only_matching': True,
4767 }, {
4768 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
4769 'only_matching': True,
4770 }, {
4771 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
4772 'only_matching': True,
4773 }, {
4774 'note': 'A channel that is not live. Should raise error',
4775 'url': 'https://www.youtube.com/user/numberphile/live',
4776 'only_matching': True,
4777 }, {
4778 'url': 'https://www.youtube.com/feed/trending',
4779 'only_matching': True,
4780 }, {
4781 'url': 'https://www.youtube.com/feed/library',
4782 'only_matching': True,
4783 }, {
4784 'url': 'https://www.youtube.com/feed/history',
4785 'only_matching': True,
4786 }, {
4787 'url': 'https://www.youtube.com/feed/subscriptions',
4788 'only_matching': True,
4789 }, {
4790 'url': 'https://www.youtube.com/feed/watch_later',
4791 'only_matching': True,
4792 }, {
4793 'note': 'Recommended - redirects to home page.',
4794 'url': 'https://www.youtube.com/feed/recommended',
4795 'only_matching': True,
4796 }, {
4797 'note': 'inline playlist with not always working continuations',
4798 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
4799 'only_matching': True,
4800 }, {
4801 'url': 'https://www.youtube.com/course',
4802 'only_matching': True,
4803 }, {
4804 'url': 'https://www.youtube.com/zsecurity',
4805 'only_matching': True,
4806 }, {
4807 'url': 'http://www.youtube.com/NASAgovVideo/videos',
4808 'only_matching': True,
4809 }, {
4810 'url': 'https://www.youtube.com/TheYoungTurks/live',
4811 'only_matching': True,
4812 }, {
4813 'url': 'https://www.youtube.com/hashtag/cctv9',
4814 'info_dict': {
4815 'id': 'cctv9',
4816 'title': '#cctv9',
4817 'tags': [],
4818 },
4819 'playlist_mincount': 350,
4820 }, {
4821 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
4822 'only_matching': True,
4823 }, {
4824 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
4825 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
4826 'only_matching': True
4827 }, {
4828 'note': '/browse/ should redirect to /channel/',
4829 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
4830 'only_matching': True
4831 }, {
4832 'note': 'VLPL, should redirect to playlist?list=PL...',
4833 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
4834 'info_dict': {
4835 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
4836 'uploader': 'NoCopyrightSounds',
4837 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
4838 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
4839 'title': 'NCS Releases',
4840 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
4841 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
4842 'modified_date': r're:\d{8}',
4843 'view_count': int,
4844 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
4845 'tags': [],
4846 'channel': 'NoCopyrightSounds',
4847 },
4848 'playlist_mincount': 166,
4849 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4850 }, {
4851 'note': 'Topic, should redirect to playlist?list=UU...',
4852 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
4853 'info_dict': {
4854 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
4855 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
4856 'title': 'Uploads from Royalty Free Music - Topic',
4857 'uploader': 'Royalty Free Music - Topic',
4858 'tags': [],
4859 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
4860 'channel': 'Royalty Free Music - Topic',
4861 'view_count': int,
4862 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
4863 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
4864 'modified_date': r're:\d{8}',
4865 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
4866 'description': '',
4867 },
4868 'expected_warnings': [
4869 'The URL does not have a videos tab',
4870 r'[Uu]navailable videos (are|will be) hidden',
4871 ],
4872 'playlist_mincount': 101,
4873 }, {
4874 'note': 'Topic without a UU playlist',
4875 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
4876 'info_dict': {
4877 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
4878 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
4879 'tags': [],
4880 },
4881 'expected_warnings': [
4882 'the playlist redirect gave error',
4883 ],
4884 'playlist_mincount': 9,
4885 }, {
4886 'note': 'Youtube music Album',
4887 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
4888 'info_dict': {
4889 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
4890 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
4891 'tags': [],
4892 'view_count': int,
4893 'description': '',
4894 'availability': 'unlisted',
4895 'modified_date': r're:\d{8}',
4896 },
4897 'playlist_count': 50,
4898 }, {
4899 'note': 'unlisted single video playlist',
4900 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
4901 'info_dict': {
4902 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
4903 'uploader': 'colethedj',
4904 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
4905 'title': 'yt-dlp unlisted playlist test',
4906 'availability': 'unlisted',
4907 'tags': [],
4908 'modified_date': '20211208',
4909 'channel': 'colethedj',
4910 'view_count': int,
4911 'description': '',
4912 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
4913 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
4914 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
4915 },
4916 'playlist_count': 1,
4917 }, {
4918 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
4919 'url': 'https://www.youtube.com/feed/recommended',
4920 'info_dict': {
4921 'id': 'recommended',
4922 'title': 'recommended',
4923 'tags': [],
4924 },
4925 'playlist_mincount': 50,
4926 'params': {
4927 'skip_download': True,
4928 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
4929 },
4930 }, {
4931 'note': 'API Fallback: /videos tab, sorted by oldest first',
4932 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
4933 'info_dict': {
4934 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
4935 'title': 'Cody\'sLab - Videos',
4936 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
4937 'uploader': 'Cody\'sLab',
4938 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
4939 'channel': 'Cody\'sLab',
4940 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
4941 'tags': [],
4942 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
4943 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
4944 'channel_follower_count': int
4945 },
4946 'playlist_mincount': 650,
4947 'params': {
4948 'skip_download': True,
4949 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
4950 },
4951 }, {
4952 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
4953 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
4954 'info_dict': {
4955 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
4956 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
4957 'title': 'Uploads from Royalty Free Music - Topic',
4958 'uploader': 'Royalty Free Music - Topic',
4959 'modified_date': r're:\d{8}',
4960 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
4961 'description': '',
4962 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
4963 'tags': [],
4964 'channel': 'Royalty Free Music - Topic',
4965 'view_count': int,
4966 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
4967 },
4968 'expected_warnings': [
4969 'does not have a videos tab',
4970 r'[Uu]navailable videos (are|will be) hidden',
4971 ],
4972 'playlist_mincount': 101,
4973 'params': {
4974 'skip_download': True,
4975 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
4976 },
4977 }]
4978
4979 @classmethod
4980 def suitable(cls, url):
4981 return False if YoutubeIE.suitable(url) else super(
4982 YoutubeTabIE, cls).suitable(url)
4983
4984 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
4985
4986 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
4987 def _real_extract(self, url, smuggled_data):
4988 item_id = self._match_id(url)
4989 url = compat_urlparse.urlunparse(
4990 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
4991 compat_opts = self.get_param('compat_opts', [])
4992
4993 def get_mobj(url):
4994 mobj = self._URL_RE.match(url).groupdict()
4995 mobj.update((k, '') for k, v in mobj.items() if v is None)
4996 return mobj
4997
4998 mobj, redirect_warning = get_mobj(url), None
4999 # Youtube returns incomplete data if tabname is not lower case
5000 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5001 if is_channel:
5002 if smuggled_data.get('is_music_url'):
5003 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5004 item_id = item_id[2:]
5005 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5006 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5007 mdata = self._extract_tab_endpoint(
5008 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5009 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5010 get_all=False, expected_type=compat_str)
5011 if not murl:
5012 raise ExtractorError('Failed to resolve album to playlist')
5013 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5014 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5015 pre = f'https://www.youtube.com/channel/{item_id}'
5016
5017 original_tab_name = tab
5018 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5019 # Home URLs should redirect to /videos/
5020 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5021 'To download only the videos in the home page, add a "/featured" to the URL')
5022 tab = '/videos'
5023
5024 url = ''.join((pre, tab, post))
5025 mobj = get_mobj(url)
5026
5027 # Handle both video/playlist URLs
5028 qs = parse_qs(url)
5029 video_id, playlist_id = [qs.get(key, [None])[0] for key in ('v', 'list')]
5030
5031 if not video_id and mobj['not_channel'].startswith('watch'):
5032 if not playlist_id:
5033 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5034 raise ExtractorError('Unable to recognize tab page')
5035 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5036 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5037 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5038 mobj = get_mobj(url)
5039
5040 if video_id and playlist_id:
5041 if self.get_param('noplaylist'):
5042 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5043 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5044 ie=YoutubeIE.ie_key(), video_id=video_id)
5045 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5046
5047 data, ytcfg = self._extract_data(url, item_id)
5048
5049 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5050 if tabs:
5051 selected_tab = self._extract_selected_tab(tabs)
5052 selected_tab_name = selected_tab.get('title', '').lower()
5053 if selected_tab_name == 'home':
5054 selected_tab_name = 'featured'
5055 requested_tab_name = mobj['tab'][1:]
5056 if 'no-youtube-channel-redirect' not in compat_opts:
5057 if requested_tab_name == 'live':
5058 # Live tab should have redirected to the video
5059 raise ExtractorError('The channel is not currently live', expected=True)
5060 if requested_tab_name not in ('', selected_tab_name):
5061 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5062 if not original_tab_name:
5063 if item_id[:2] == 'UC':
5064 # Topic channels don't have /videos. Use the equivalent playlist instead
5065 pl_id = f'UU{item_id[2:]}'
5066 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5067 try:
5068 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5069 except ExtractorError:
5070 redirect_warning += ' and the playlist redirect gave error'
5071 else:
5072 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5073 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5074 if selected_tab_name and selected_tab_name != requested_tab_name:
5075 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5076 else:
5077 raise ExtractorError(redirect_warning, expected=True)
5078
5079 if redirect_warning:
5080 self.to_screen(redirect_warning)
5081 self.write_debug(f'Final URL: {url}')
5082
5083 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5084 if 'no-youtube-unavailable-videos' not in compat_opts:
5085 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5086 self._extract_and_report_alerts(data, only_once=True)
5087 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5088 if tabs:
5089 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5090
5091 playlist = traverse_obj(
5092 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5093 if playlist:
5094 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5095
5096 video_id = traverse_obj(
5097 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5098 if video_id:
5099 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5100 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5101 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5102 ie=YoutubeIE.ie_key(), video_id=video_id)
5103
5104 raise ExtractorError('Unable to recognize tab page')
5105
5106
5107 class YoutubePlaylistIE(InfoExtractor):
5108 IE_DESC = 'YouTube playlists'
5109 _VALID_URL = r'''(?x)(?:
5110 (?:https?://)?
5111 (?:\w+\.)?
5112 (?:
5113 (?:
5114 youtube(?:kids)?\.com|
5115 %(invidious)s
5116 )
5117 /.*?\?.*?\blist=
5118 )?
5119 (?P<id>%(playlist_id)s)
5120 )''' % {
5121 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5122 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5123 }
5124 IE_NAME = 'youtube:playlist'
5125 _TESTS = [{
5126 'note': 'issue #673',
5127 'url': 'PLBB231211A4F62143',
5128 'info_dict': {
5129 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5130 'id': 'PLBB231211A4F62143',
5131 'uploader': 'Wickman',
5132 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5133 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5134 'view_count': int,
5135 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5136 'modified_date': r're:\d{8}',
5137 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5138 'channel': 'Wickman',
5139 'tags': [],
5140 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5141 },
5142 'playlist_mincount': 29,
5143 }, {
5144 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5145 'info_dict': {
5146 'title': 'YDL_safe_search',
5147 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5148 },
5149 'playlist_count': 2,
5150 'skip': 'This playlist is private',
5151 }, {
5152 'note': 'embedded',
5153 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5154 'playlist_count': 4,
5155 'info_dict': {
5156 'title': 'JODA15',
5157 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5158 'uploader': 'milan',
5159 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5160 'description': '',
5161 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5162 'tags': [],
5163 'modified_date': '20140919',
5164 'view_count': int,
5165 'channel': 'milan',
5166 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5167 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5168 },
5169 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5170 }, {
5171 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5172 'playlist_mincount': 654,
5173 'info_dict': {
5174 'title': '2018 Chinese New Singles (11/6 updated)',
5175 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5176 'uploader': 'LBK',
5177 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5178 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5179 'channel': 'LBK',
5180 'view_count': int,
5181 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5182 'tags': [],
5183 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5184 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5185 'modified_date': r're:\d{8}',
5186 },
5187 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5188 }, {
5189 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5190 'only_matching': True,
5191 }, {
5192 # music album playlist
5193 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5194 'only_matching': True,
5195 }]
5196
5197 @classmethod
5198 def suitable(cls, url):
5199 if YoutubeTabIE.suitable(url):
5200 return False
5201 from ..utils import parse_qs
5202 qs = parse_qs(url)
5203 if qs.get('v', [None])[0]:
5204 return False
5205 return super(YoutubePlaylistIE, cls).suitable(url)
5206
5207 def _real_extract(self, url):
5208 playlist_id = self._match_id(url)
5209 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5210 url = update_url_query(
5211 'https://www.youtube.com/playlist',
5212 parse_qs(url) or {'list': playlist_id})
5213 if is_music_url:
5214 url = smuggle_url(url, {'is_music_url': True})
5215 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5216
5217
5218 class YoutubeYtBeIE(InfoExtractor):
5219 IE_DESC = 'youtu.be'
5220 _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}
5221 _TESTS = [{
5222 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5223 'info_dict': {
5224 'id': 'yeWKywCrFtk',
5225 'ext': 'mp4',
5226 'title': 'Small Scale Baler and Braiding Rugs',
5227 'uploader': 'Backus-Page House Museum',
5228 'uploader_id': 'backuspagemuseum',
5229 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5230 'upload_date': '20161008',
5231 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5232 'categories': ['Nonprofits & Activism'],
5233 'tags': list,
5234 'like_count': int,
5235 'age_limit': 0,
5236 'playable_in_embed': True,
5237 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5238 'channel': 'Backus-Page House Museum',
5239 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5240 'live_status': 'not_live',
5241 'view_count': int,
5242 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5243 'availability': 'public',
5244 'duration': 59,
5245 },
5246 'params': {
5247 'noplaylist': True,
5248 'skip_download': True,
5249 },
5250 }, {
5251 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5252 'only_matching': True,
5253 }]
5254
5255 def _real_extract(self, url):
5256 mobj = self._match_valid_url(url)
5257 video_id = mobj.group('id')
5258 playlist_id = mobj.group('playlist_id')
5259 return self.url_result(
5260 update_url_query('https://www.youtube.com/watch', {
5261 'v': video_id,
5262 'list': playlist_id,
5263 'feature': 'youtu.be',
5264 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5265
5266
5267 class YoutubeLivestreamEmbedIE(InfoExtractor):
5268 IE_DESC = 'YouTube livestream embeds'
5269 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5270 _TESTS = [{
5271 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5272 'only_matching': True,
5273 }]
5274
5275 def _real_extract(self, url):
5276 channel_id = self._match_id(url)
5277 return self.url_result(
5278 f'https://www.youtube.com/channel/{channel_id}/live',
5279 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5280
5281
5282 class YoutubeYtUserIE(InfoExtractor):
5283 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5284 IE_NAME = 'youtube:user'
5285 _VALID_URL = r'ytuser:(?P<id>.+)'
5286 _TESTS = [{
5287 'url': 'ytuser:phihag',
5288 'only_matching': True,
5289 }]
5290
5291 def _real_extract(self, url):
5292 user_id = self._match_id(url)
5293 return self.url_result(
5294 'https://www.youtube.com/user/%s/videos' % user_id,
5295 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5296
5297
5298 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5299 IE_NAME = 'youtube:favorites'
5300 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5301 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5302 _LOGIN_REQUIRED = True
5303 _TESTS = [{
5304 'url': ':ytfav',
5305 'only_matching': True,
5306 }, {
5307 'url': ':ytfavorites',
5308 'only_matching': True,
5309 }]
5310
5311 def _real_extract(self, url):
5312 return self.url_result(
5313 'https://www.youtube.com/playlist?list=LL',
5314 ie=YoutubeTabIE.ie_key())
5315
5316
5317 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5318 IE_DESC = 'YouTube search'
5319 IE_NAME = 'youtube:search'
5320 _SEARCH_KEY = 'ytsearch'
5321 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5322 _TESTS = [{
5323 'url': 'ytsearch5:youtube-dl test video',
5324 'playlist_count': 5,
5325 'info_dict': {
5326 'id': 'youtube-dl test video',
5327 'title': 'youtube-dl test video',
5328 }
5329 }]
5330
5331
5332 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5333 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5334 _SEARCH_KEY = 'ytsearchdate'
5335 IE_DESC = 'YouTube search, newest videos first'
5336 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5337 _TESTS = [{
5338 'url': 'ytsearchdate5:youtube-dl test video',
5339 'playlist_count': 5,
5340 'info_dict': {
5341 'id': 'youtube-dl test video',
5342 'title': 'youtube-dl test video',
5343 }
5344 }]
5345
5346
5347 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5348 IE_DESC = 'YouTube search URLs with sorting and filter support'
5349 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5350 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5351 _TESTS = [{
5352 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5353 'playlist_mincount': 5,
5354 'info_dict': {
5355 'id': 'youtube-dl test video',
5356 'title': 'youtube-dl test video',
5357 }
5358 }, {
5359 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5360 'playlist_mincount': 5,
5361 'info_dict': {
5362 'id': 'python',
5363 'title': 'python',
5364 }
5365
5366 }, {
5367 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5368 'only_matching': True,
5369 }]
5370
5371 def _real_extract(self, url):
5372 qs = parse_qs(url)
5373 query = (qs.get('search_query') or qs.get('q'))[0]
5374 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5375
5376
5377 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5378 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5379 IE_NAME = 'youtube:music:search_url'
5380 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5381 _TESTS = [{
5382 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5383 'playlist_count': 16,
5384 'info_dict': {
5385 'id': 'royalty free music',
5386 'title': 'royalty free music',
5387 }
5388 }, {
5389 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5390 'playlist_mincount': 30,
5391 'info_dict': {
5392 'id': 'royalty free music - songs',
5393 'title': 'royalty free music - songs',
5394 },
5395 'params': {'extract_flat': 'in_playlist'}
5396 }, {
5397 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5398 'playlist_mincount': 30,
5399 'info_dict': {
5400 'id': 'royalty free music - community playlists',
5401 'title': 'royalty free music - community playlists',
5402 },
5403 'params': {'extract_flat': 'in_playlist'}
5404 }]
5405
5406 _SECTIONS = {
5407 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5408 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5409 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5410 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5411 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5412 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5413 }
5414
5415 def _real_extract(self, url):
5416 qs = parse_qs(url)
5417 query = (qs.get('search_query') or qs.get('q'))[0]
5418 params = qs.get('sp', (None,))[0]
5419 if params:
5420 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5421 else:
5422 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5423 params = self._SECTIONS.get(section)
5424 if not params:
5425 section = None
5426 title = join_nonempty(query, section, delim=' - ')
5427 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5428
5429
5430 class YoutubeFeedsInfoExtractor(InfoExtractor):
5431 """
5432 Base class for feed extractors
5433 Subclasses must define the _FEED_NAME property.
5434 """
5435 _LOGIN_REQUIRED = True
5436 _TESTS = []
5437
5438 @property
5439 def IE_NAME(self):
5440 return 'youtube:%s' % self._FEED_NAME
5441
5442 def _real_extract(self, url):
5443 return self.url_result(
5444 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5445
5446
5447 class YoutubeWatchLaterIE(InfoExtractor):
5448 IE_NAME = 'youtube:watchlater'
5449 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5450 _VALID_URL = r':ytwatchlater'
5451 _TESTS = [{
5452 'url': ':ytwatchlater',
5453 'only_matching': True,
5454 }]
5455
5456 def _real_extract(self, url):
5457 return self.url_result(
5458 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5459
5460
5461 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5462 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5463 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5464 _FEED_NAME = 'recommended'
5465 _LOGIN_REQUIRED = False
5466 _TESTS = [{
5467 'url': ':ytrec',
5468 'only_matching': True,
5469 }, {
5470 'url': ':ytrecommended',
5471 'only_matching': True,
5472 }, {
5473 'url': 'https://youtube.com',
5474 'only_matching': True,
5475 }]
5476
5477
5478 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5479 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5480 _VALID_URL = r':ytsub(?:scription)?s?'
5481 _FEED_NAME = 'subscriptions'
5482 _TESTS = [{
5483 'url': ':ytsubs',
5484 'only_matching': True,
5485 }, {
5486 'url': ':ytsubscriptions',
5487 'only_matching': True,
5488 }]
5489
5490
5491 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5492 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5493 _VALID_URL = r':ythis(?:tory)?'
5494 _FEED_NAME = 'history'
5495 _TESTS = [{
5496 'url': ':ythistory',
5497 'only_matching': True,
5498 }]
5499
5500
5501 class YoutubeTruncatedURLIE(InfoExtractor):
5502 IE_NAME = 'youtube:truncated_url'
5503 IE_DESC = False # Do not list
5504 _VALID_URL = r'''(?x)
5505 (?:https?://)?
5506 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5507 (?:watch\?(?:
5508 feature=[a-z_]+|
5509 annotation_id=annotation_[^&]+|
5510 x-yt-cl=[0-9]+|
5511 hl=[^&]*|
5512 t=[0-9]+
5513 )?
5514 |
5515 attribution_link\?a=[^&]+
5516 )
5517 $
5518 '''
5519
5520 _TESTS = [{
5521 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5522 'only_matching': True,
5523 }, {
5524 'url': 'https://www.youtube.com/watch?',
5525 'only_matching': True,
5526 }, {
5527 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5528 'only_matching': True,
5529 }, {
5530 'url': 'https://www.youtube.com/watch?feature=foo',
5531 'only_matching': True,
5532 }, {
5533 'url': 'https://www.youtube.com/watch?hl=en-GB',
5534 'only_matching': True,
5535 }, {
5536 'url': 'https://www.youtube.com/watch?t=2372',
5537 'only_matching': True,
5538 }]
5539
5540 def _real_extract(self, url):
5541 raise ExtractorError(
5542 'Did you forget to quote the URL? Remember that & is a meta '
5543 'character in most shells, so you want to put the URL in quotes, '
5544 'like youtube-dl '
5545 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5546 ' or simply youtube-dl BaW_jenozKc .',
5547 expected=True)
5548
5549
5550 class YoutubeClipIE(InfoExtractor):
5551 IE_NAME = 'youtube:clip'
5552 IE_DESC = False # Do not list
5553 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5554
5555 def _real_extract(self, url):
5556 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5557 return self.url_result(url, 'Generic')
5558
5559
5560 class YoutubeTruncatedIDIE(InfoExtractor):
5561 IE_NAME = 'youtube:truncated_id'
5562 IE_DESC = False # Do not list
5563 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5564
5565 _TESTS = [{
5566 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5567 'only_matching': True,
5568 }]
5569
5570 def _real_extract(self, url):
5571 video_id = self._match_id(url)
5572 raise ExtractorError(
5573 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
5574 expected=True)