]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
345fc1e93b31efc365e427c116ee473bdb552209
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import copy
4 import datetime
5 import functools
6 import hashlib
7 import itertools
8 import json
9 import math
10 import os.path
11 import random
12 import re
13 import sys
14 import threading
15 import time
16 import traceback
17
18 from .common import InfoExtractor, SearchInfoExtractor
19 from ..compat import (
20 compat_chr,
21 compat_HTTPError,
22 compat_parse_qs,
23 compat_str,
24 compat_urllib_parse_unquote_plus,
25 compat_urllib_parse_urlencode,
26 compat_urllib_parse_urlparse,
27 compat_urlparse,
28 )
29 from ..jsinterp import JSInterpreter
30 from ..utils import (
31 NO_DEFAULT,
32 ExtractorError,
33 bug_reports_message,
34 classproperty,
35 clean_html,
36 datetime_from_str,
37 dict_get,
38 error_to_compat_str,
39 float_or_none,
40 format_field,
41 get_first,
42 int_or_none,
43 is_html,
44 join_nonempty,
45 js_to_json,
46 mimetype2ext,
47 network_exceptions,
48 orderedSet,
49 parse_codecs,
50 parse_count,
51 parse_duration,
52 parse_iso8601,
53 parse_qs,
54 qualities,
55 remove_end,
56 remove_start,
57 smuggle_url,
58 str_or_none,
59 str_to_int,
60 strftime_or_none,
61 traverse_obj,
62 try_get,
63 unescapeHTML,
64 unified_strdate,
65 unified_timestamp,
66 unsmuggle_url,
67 update_url_query,
68 url_or_none,
69 urljoin,
70 variadic,
71 )
72
73 # any clients starting with _ cannot be explicity requested by the user
74 INNERTUBE_CLIENTS = {
75 'web': {
76 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
77 'INNERTUBE_CONTEXT': {
78 'client': {
79 'clientName': 'WEB',
80 'clientVersion': '2.20211221.00.00',
81 }
82 },
83 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
84 },
85 'web_embedded': {
86 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
87 'INNERTUBE_CONTEXT': {
88 'client': {
89 'clientName': 'WEB_EMBEDDED_PLAYER',
90 'clientVersion': '1.20211215.00.01',
91 },
92 },
93 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
94 },
95 'web_music': {
96 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
97 'INNERTUBE_HOST': 'music.youtube.com',
98 'INNERTUBE_CONTEXT': {
99 'client': {
100 'clientName': 'WEB_REMIX',
101 'clientVersion': '1.20211213.00.00',
102 }
103 },
104 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
105 },
106 'web_creator': {
107 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
108 'INNERTUBE_CONTEXT': {
109 'client': {
110 'clientName': 'WEB_CREATOR',
111 'clientVersion': '1.20211220.02.00',
112 }
113 },
114 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
115 },
116 'android': {
117 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
118 'INNERTUBE_CONTEXT': {
119 'client': {
120 'clientName': 'ANDROID',
121 'clientVersion': '16.49',
122 }
123 },
124 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
125 'REQUIRE_JS_PLAYER': False
126 },
127 'android_embedded': {
128 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
129 'INNERTUBE_CONTEXT': {
130 'client': {
131 'clientName': 'ANDROID_EMBEDDED_PLAYER',
132 'clientVersion': '16.49',
133 },
134 },
135 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
136 'REQUIRE_JS_PLAYER': False
137 },
138 'android_music': {
139 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
140 'INNERTUBE_CONTEXT': {
141 'client': {
142 'clientName': 'ANDROID_MUSIC',
143 'clientVersion': '4.57',
144 }
145 },
146 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
147 'REQUIRE_JS_PLAYER': False
148 },
149 'android_creator': {
150 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
151 'INNERTUBE_CONTEXT': {
152 'client': {
153 'clientName': 'ANDROID_CREATOR',
154 'clientVersion': '21.47',
155 },
156 },
157 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
158 'REQUIRE_JS_PLAYER': False
159 },
160 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
161 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
162 'ios': {
163 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
164 'INNERTUBE_CONTEXT': {
165 'client': {
166 'clientName': 'IOS',
167 'clientVersion': '16.46',
168 'deviceModel': 'iPhone14,3',
169 }
170 },
171 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
172 'REQUIRE_JS_PLAYER': False
173 },
174 'ios_embedded': {
175 'INNERTUBE_CONTEXT': {
176 'client': {
177 'clientName': 'IOS_MESSAGES_EXTENSION',
178 'clientVersion': '16.46',
179 'deviceModel': 'iPhone14,3',
180 },
181 },
182 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
183 'REQUIRE_JS_PLAYER': False
184 },
185 'ios_music': {
186 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
187 'INNERTUBE_CONTEXT': {
188 'client': {
189 'clientName': 'IOS_MUSIC',
190 'clientVersion': '4.57',
191 },
192 },
193 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
194 'REQUIRE_JS_PLAYER': False
195 },
196 'ios_creator': {
197 'INNERTUBE_CONTEXT': {
198 'client': {
199 'clientName': 'IOS_CREATOR',
200 'clientVersion': '21.47',
201 },
202 },
203 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
204 'REQUIRE_JS_PLAYER': False
205 },
206 # mweb has 'ultralow' formats
207 # See: https://github.com/yt-dlp/yt-dlp/pull/557
208 'mweb': {
209 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
210 'INNERTUBE_CONTEXT': {
211 'client': {
212 'clientName': 'MWEB',
213 'clientVersion': '2.20211221.01.00',
214 }
215 },
216 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
217 },
218 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
219 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
220 'tv_embedded': {
221 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
222 'INNERTUBE_CONTEXT': {
223 'client': {
224 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
225 'clientVersion': '2.0',
226 },
227 },
228 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
229 },
230 }
231
232
233 def _split_innertube_client(client_name):
234 variant, *base = client_name.rsplit('.', 1)
235 if base:
236 return variant, base[0], variant
237 base, *variant = client_name.split('_', 1)
238 return client_name, base, variant[0] if variant else None
239
240
241 def build_innertube_clients():
242 THIRD_PARTY = {
243 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
244 }
245 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
246 priority = qualities(BASE_CLIENTS[::-1])
247
248 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
249 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
250 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
251 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
252 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
253
254 _, base_client, variant = _split_innertube_client(client)
255 ytcfg['priority'] = 10 * priority(base_client)
256
257 if not variant:
258 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
259 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
260 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
261 embedscreen['priority'] -= 3
262 elif variant == 'embedded':
263 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
264 ytcfg['priority'] -= 2
265 else:
266 ytcfg['priority'] -= 3
267
268
269 build_innertube_clients()
270
271
272 class YoutubeBaseInfoExtractor(InfoExtractor):
273 """Provide base functions for Youtube extractors"""
274
275 _RESERVED_NAMES = (
276 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
277 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
278 r'browse|oembed|get_video_info|iframe_api|s/player|'
279 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
280
281 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
282
283 # _NETRC_MACHINE = 'youtube'
284
285 # If True it will raise an error if no login info is provided
286 _LOGIN_REQUIRED = False
287
288 _INVIDIOUS_SITES = (
289 # invidious-redirect websites
290 r'(?:www\.)?redirect\.invidious\.io',
291 r'(?:(?:www|dev)\.)?invidio\.us',
292 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
293 r'(?:www\.)?invidious\.pussthecat\.org',
294 r'(?:www\.)?invidious\.zee\.li',
295 r'(?:www\.)?invidious\.ethibox\.fr',
296 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
297 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
298 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
299 # youtube-dl invidious instances list
300 r'(?:(?:www|no)\.)?invidiou\.sh',
301 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
302 r'(?:www\.)?invidious\.kabi\.tk',
303 r'(?:www\.)?invidious\.mastodon\.host',
304 r'(?:www\.)?invidious\.zapashcanon\.fr',
305 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
306 r'(?:www\.)?invidious\.tinfoil-hat\.net',
307 r'(?:www\.)?invidious\.himiko\.cloud',
308 r'(?:www\.)?invidious\.reallyancient\.tech',
309 r'(?:www\.)?invidious\.tube',
310 r'(?:www\.)?invidiou\.site',
311 r'(?:www\.)?invidious\.site',
312 r'(?:www\.)?invidious\.xyz',
313 r'(?:www\.)?invidious\.nixnet\.xyz',
314 r'(?:www\.)?invidious\.048596\.xyz',
315 r'(?:www\.)?invidious\.drycat\.fr',
316 r'(?:www\.)?inv\.skyn3t\.in',
317 r'(?:www\.)?tube\.poal\.co',
318 r'(?:www\.)?tube\.connect\.cafe',
319 r'(?:www\.)?vid\.wxzm\.sx',
320 r'(?:www\.)?vid\.mint\.lgbt',
321 r'(?:www\.)?vid\.puffyan\.us',
322 r'(?:www\.)?yewtu\.be',
323 r'(?:www\.)?yt\.elukerio\.org',
324 r'(?:www\.)?yt\.lelux\.fi',
325 r'(?:www\.)?invidious\.ggc-project\.de',
326 r'(?:www\.)?yt\.maisputain\.ovh',
327 r'(?:www\.)?ytprivate\.com',
328 r'(?:www\.)?invidious\.13ad\.de',
329 r'(?:www\.)?invidious\.toot\.koeln',
330 r'(?:www\.)?invidious\.fdn\.fr',
331 r'(?:www\.)?watch\.nettohikari\.com',
332 r'(?:www\.)?invidious\.namazso\.eu',
333 r'(?:www\.)?invidious\.silkky\.cloud',
334 r'(?:www\.)?invidious\.exonip\.de',
335 r'(?:www\.)?invidious\.riverside\.rocks',
336 r'(?:www\.)?invidious\.blamefran\.net',
337 r'(?:www\.)?invidious\.moomoo\.de',
338 r'(?:www\.)?ytb\.trom\.tf',
339 r'(?:www\.)?yt\.cyberhost\.uk',
340 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
341 r'(?:www\.)?qklhadlycap4cnod\.onion',
342 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
343 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
344 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
345 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
346 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
347 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
348 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
349 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
350 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
351 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
352 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
353 r'(?:www\.)?piped\.kavin\.rocks',
354 r'(?:www\.)?piped\.silkky\.cloud',
355 r'(?:www\.)?piped\.tokhmi\.xyz',
356 r'(?:www\.)?piped\.moomoo\.me',
357 r'(?:www\.)?il\.ax',
358 r'(?:www\.)?piped\.syncpundit\.com',
359 r'(?:www\.)?piped\.mha\.fi',
360 r'(?:www\.)?piped\.mint\.lgbt',
361 r'(?:www\.)?piped\.privacy\.com\.de',
362 )
363
364 def _initialize_consent(self):
365 cookies = self._get_cookies('https://www.youtube.com/')
366 if cookies.get('__Secure-3PSID'):
367 return
368 consent_id = None
369 consent = cookies.get('CONSENT')
370 if consent:
371 if 'YES' in consent.value:
372 return
373 consent_id = self._search_regex(
374 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
375 if not consent_id:
376 consent_id = random.randint(100, 999)
377 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
378
379 def _initialize_pref(self):
380 cookies = self._get_cookies('https://www.youtube.com/')
381 pref_cookie = cookies.get('PREF')
382 pref = {}
383 if pref_cookie:
384 try:
385 pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
386 except ValueError:
387 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
388 pref.update({'hl': 'en', 'tz': 'UTC'})
389 self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
390
391 def _real_initialize(self):
392 self._initialize_pref()
393 self._initialize_consent()
394 self._check_login_required()
395
396 def _check_login_required(self):
397 if (self._LOGIN_REQUIRED
398 and self.get_param('cookiefile') is None
399 and self.get_param('cookiesfrombrowser') is None):
400 self.raise_login_required('Login details are needed to download this content', method='cookies')
401
402 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
403 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
404 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
405
406 def _get_default_ytcfg(self, client='web'):
407 return copy.deepcopy(INNERTUBE_CLIENTS[client])
408
409 def _get_innertube_host(self, client='web'):
410 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
411
412 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
413 # try_get but with fallback to default ytcfg client values when present
414 _func = lambda y: try_get(y, getter, expected_type)
415 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
416
417 def _extract_client_name(self, ytcfg, default_client='web'):
418 return self._ytcfg_get_safe(
419 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
420 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
421
422 def _extract_client_version(self, ytcfg, default_client='web'):
423 return self._ytcfg_get_safe(
424 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
425 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
426
427 def _extract_api_key(self, ytcfg=None, default_client='web'):
428 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
429
430 def _extract_context(self, ytcfg=None, default_client='web'):
431 context = get_first(
432 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
433 # Enforce language and tz for extraction
434 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
435 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
436 return context
437
438 _SAPISID = None
439
440 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
441 time_now = round(time.time())
442 if self._SAPISID is None:
443 yt_cookies = self._get_cookies('https://www.youtube.com')
444 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
445 # See: https://github.com/yt-dlp/yt-dlp/issues/393
446 sapisid_cookie = dict_get(
447 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
448 if sapisid_cookie and sapisid_cookie.value:
449 self._SAPISID = sapisid_cookie.value
450 self.write_debug('Extracted SAPISID cookie')
451 # SAPISID cookie is required if not already present
452 if not yt_cookies.get('SAPISID'):
453 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
454 self._set_cookie(
455 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
456 else:
457 self._SAPISID = False
458 if not self._SAPISID:
459 return None
460 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
461 sapisidhash = hashlib.sha1(
462 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
463 return f'SAPISIDHASH {time_now}_{sapisidhash}'
464
465 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
466 note='Downloading API JSON', errnote='Unable to download API page',
467 context=None, api_key=None, api_hostname=None, default_client='web'):
468
469 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
470 data.update(query)
471 real_headers = self.generate_api_headers(default_client=default_client)
472 real_headers.update({'content-type': 'application/json'})
473 if headers:
474 real_headers.update(headers)
475 return self._download_json(
476 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
477 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
478 data=json.dumps(data).encode('utf8'), headers=real_headers,
479 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
480
481 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
482 data = self._search_regex(
483 (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
484 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
485 if data:
486 return self._parse_json(data, item_id, fatal=fatal)
487
488 @staticmethod
489 def _extract_session_index(*data):
490 """
491 Index of current account in account list.
492 See: https://github.com/yt-dlp/yt-dlp/pull/519
493 """
494 for ytcfg in data:
495 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
496 if session_index is not None:
497 return session_index
498
499 # Deprecated?
500 def _extract_identity_token(self, ytcfg=None, webpage=None):
501 if ytcfg:
502 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
503 if token:
504 return token
505 if webpage:
506 return self._search_regex(
507 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
508 'identity token', default=None, fatal=False)
509
510 @staticmethod
511 def _extract_account_syncid(*args):
512 """
513 Extract syncId required to download private playlists of secondary channels
514 @params response and/or ytcfg
515 """
516 for data in args:
517 # ytcfg includes channel_syncid if on secondary channel
518 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
519 if delegated_sid:
520 return delegated_sid
521 sync_ids = (try_get(
522 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
523 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
524 if len(sync_ids) >= 2 and sync_ids[1]:
525 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
526 # and just "user_syncid||" for primary channel. We only want the channel_syncid
527 return sync_ids[0]
528
529 @staticmethod
530 def _extract_visitor_data(*args):
531 """
532 Extracts visitorData from an API response or ytcfg
533 Appears to be used to track session state
534 """
535 return get_first(
536 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
537 expected_type=str)
538
539 @property
540 def is_authenticated(self):
541 return bool(self._generate_sapisidhash_header())
542
543 def extract_ytcfg(self, video_id, webpage):
544 if not webpage:
545 return {}
546 return self._parse_json(
547 self._search_regex(
548 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
549 default='{}'), video_id, fatal=False) or {}
550
551 def generate_api_headers(
552 self, *, ytcfg=None, account_syncid=None, session_index=None,
553 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
554
555 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
556 headers = {
557 'X-YouTube-Client-Name': compat_str(
558 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
559 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
560 'Origin': origin,
561 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
562 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
563 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
564 }
565 if session_index is None:
566 session_index = self._extract_session_index(ytcfg)
567 if account_syncid or session_index is not None:
568 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
569
570 auth = self._generate_sapisidhash_header(origin)
571 if auth is not None:
572 headers['Authorization'] = auth
573 headers['X-Origin'] = origin
574 return {h: v for h, v in headers.items() if v is not None}
575
576 def _download_ytcfg(self, client, video_id):
577 url = {
578 'web': 'https://www.youtube.com',
579 'web_music': 'https://music.youtube.com',
580 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
581 }.get(client)
582 if not url:
583 return {}
584 webpage = self._download_webpage(
585 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
586 return self.extract_ytcfg(video_id, webpage) or {}
587
588 @staticmethod
589 def _build_api_continuation_query(continuation, ctp=None):
590 query = {
591 'continuation': continuation
592 }
593 # TODO: Inconsistency with clickTrackingParams.
594 # Currently we have a fixed ctp contained within context (from ytcfg)
595 # and a ctp in root query for continuation.
596 if ctp:
597 query['clickTracking'] = {'clickTrackingParams': ctp}
598 return query
599
600 @classmethod
601 def _extract_next_continuation_data(cls, renderer):
602 next_continuation = try_get(
603 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
604 lambda x: x['continuation']['reloadContinuationData']), dict)
605 if not next_continuation:
606 return
607 continuation = next_continuation.get('continuation')
608 if not continuation:
609 return
610 ctp = next_continuation.get('clickTrackingParams')
611 return cls._build_api_continuation_query(continuation, ctp)
612
613 @classmethod
614 def _extract_continuation_ep_data(cls, continuation_ep: dict):
615 if isinstance(continuation_ep, dict):
616 continuation = try_get(
617 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
618 if not continuation:
619 return
620 ctp = continuation_ep.get('clickTrackingParams')
621 return cls._build_api_continuation_query(continuation, ctp)
622
623 @classmethod
624 def _extract_continuation(cls, renderer):
625 next_continuation = cls._extract_next_continuation_data(renderer)
626 if next_continuation:
627 return next_continuation
628
629 contents = []
630 for key in ('contents', 'items'):
631 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
632
633 for content in contents:
634 if not isinstance(content, dict):
635 continue
636 continuation_ep = try_get(
637 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
638 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
639 dict)
640 continuation = cls._extract_continuation_ep_data(continuation_ep)
641 if continuation:
642 return continuation
643
644 @classmethod
645 def _extract_alerts(cls, data):
646 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
647 if not isinstance(alert_dict, dict):
648 continue
649 for alert in alert_dict.values():
650 alert_type = alert.get('type')
651 if not alert_type:
652 continue
653 message = cls._get_text(alert, 'text')
654 if message:
655 yield alert_type, message
656
657 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
658 errors = []
659 warnings = []
660 for alert_type, alert_message in alerts:
661 if alert_type.lower() == 'error' and fatal:
662 errors.append([alert_type, alert_message])
663 else:
664 warnings.append([alert_type, alert_message])
665
666 for alert_type, alert_message in (warnings + errors[:-1]):
667 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
668 if errors:
669 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
670
671 def _extract_and_report_alerts(self, data, *args, **kwargs):
672 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
673
674 def _extract_badges(self, renderer: dict):
675 badges = set()
676 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
677 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
678 if label:
679 badges.add(label.lower())
680 return badges
681
682 @staticmethod
683 def _get_text(data, *path_list, max_runs=None):
684 for path in path_list or [None]:
685 if path is None:
686 obj = [data]
687 else:
688 obj = traverse_obj(data, path, default=[])
689 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
690 obj = [obj]
691 for item in obj:
692 text = try_get(item, lambda x: x['simpleText'], compat_str)
693 if text:
694 return text
695 runs = try_get(item, lambda x: x['runs'], list) or []
696 if not runs and isinstance(item, list):
697 runs = item
698
699 runs = runs[:min(len(runs), max_runs or len(runs))]
700 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
701 if text:
702 return text
703
704 def _get_count(self, data, *path_list):
705 count_text = self._get_text(data, *path_list) or ''
706 count = parse_count(count_text)
707 if count is None:
708 count = str_to_int(
709 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
710 return count
711
712 @staticmethod
713 def _extract_thumbnails(data, *path_list):
714 """
715 Extract thumbnails from thumbnails dict
716 @param path_list: path list to level that contains 'thumbnails' key
717 """
718 thumbnails = []
719 for path in path_list or [()]:
720 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
721 thumbnail_url = url_or_none(thumbnail.get('url'))
722 if not thumbnail_url:
723 continue
724 # Sometimes youtube gives a wrong thumbnail URL. See:
725 # https://github.com/yt-dlp/yt-dlp/issues/233
726 # https://github.com/ytdl-org/youtube-dl/issues/28023
727 if 'maxresdefault' in thumbnail_url:
728 thumbnail_url = thumbnail_url.split('?')[0]
729 thumbnails.append({
730 'url': thumbnail_url,
731 'height': int_or_none(thumbnail.get('height')),
732 'width': int_or_none(thumbnail.get('width')),
733 })
734 return thumbnails
735
736 @staticmethod
737 def extract_relative_time(relative_time_text):
738 """
739 Extracts a relative time from string and converts to dt object
740 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
741 """
742 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)
743 if mobj:
744 start = mobj.group('start')
745 if start:
746 return datetime_from_str(start)
747 try:
748 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
749 except ValueError:
750 return None
751
752 def _extract_time_text(self, renderer, *path_list):
753 """@returns (timestamp, time_text)"""
754 text = self._get_text(renderer, *path_list) or ''
755 dt = self.extract_relative_time(text)
756 timestamp = None
757 if isinstance(dt, datetime.datetime):
758 timestamp = calendar.timegm(dt.timetuple())
759
760 if timestamp is None:
761 timestamp = (
762 unified_timestamp(text) or unified_timestamp(
763 self._search_regex(
764 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
765 text.lower(), 'time text', default=None)))
766
767 if text and timestamp is None:
768 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
769 return timestamp, text
770
771 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
772 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
773 default_client='web'):
774 response = None
775 last_error = None
776 count = -1
777 retries = self.get_param('extractor_retries', 3)
778 if check_get_keys is None:
779 check_get_keys = []
780 while count < retries:
781 count += 1
782 if last_error:
783 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
784 try:
785 response = self._call_api(
786 ep=ep, fatal=True, headers=headers,
787 video_id=item_id, query=query,
788 context=self._extract_context(ytcfg, default_client),
789 api_key=self._extract_api_key(ytcfg, default_client),
790 api_hostname=api_hostname, default_client=default_client,
791 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
792 except ExtractorError as e:
793 if isinstance(e.cause, network_exceptions):
794 if isinstance(e.cause, compat_HTTPError):
795 first_bytes = e.cause.read(512)
796 if not is_html(first_bytes):
797 yt_error = try_get(
798 self._parse_json(
799 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
800 lambda x: x['error']['message'], compat_str)
801 if yt_error:
802 self._report_alerts([('ERROR', yt_error)], fatal=False)
803 # Downloading page may result in intermittent 5xx HTTP error
804 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
805 # We also want to catch all other network exceptions since errors in later pages can be troublesome
806 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
807 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
808 last_error = error_to_compat_str(e.cause or e.msg)
809 if count < retries:
810 continue
811 if fatal:
812 raise
813 else:
814 self.report_warning(error_to_compat_str(e))
815 return
816
817 else:
818 try:
819 self._extract_and_report_alerts(response, only_once=True)
820 except ExtractorError as e:
821 # YouTube servers may return errors we want to retry on in a 200 OK response
822 # See: https://github.com/yt-dlp/yt-dlp/issues/839
823 if 'unknown error' in e.msg.lower():
824 last_error = e.msg
825 continue
826 if fatal:
827 raise
828 self.report_warning(error_to_compat_str(e))
829 return
830 if not check_get_keys or dict_get(response, check_get_keys):
831 break
832 # Youtube sometimes sends incomplete data
833 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
834 last_error = 'Incomplete data received'
835 if count >= retries:
836 if fatal:
837 raise ExtractorError(last_error)
838 else:
839 self.report_warning(last_error)
840 return
841 return response
842
843 @staticmethod
844 def is_music_url(url):
845 return re.match(r'https?://music\.youtube\.com/', url) is not None
846
847 def _extract_video(self, renderer):
848 video_id = renderer.get('videoId')
849 title = self._get_text(renderer, 'title')
850 description = self._get_text(renderer, 'descriptionSnippet')
851 duration = parse_duration(self._get_text(
852 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
853 if duration is None:
854 duration = parse_duration(self._search_regex(
855 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
856 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
857 video_id, default=None, group='duration'))
858
859 view_count = self._get_count(renderer, 'viewCountText')
860
861 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
862 channel_id = traverse_obj(
863 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
864 expected_type=str, get_all=False)
865 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
866 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
867 overlay_style = traverse_obj(
868 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
869 get_all=False, expected_type=str)
870 badges = self._extract_badges(renderer)
871 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
872 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
873 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
874 expected_type=str)) or ''
875 url = f'https://www.youtube.com/watch?v={video_id}'
876 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
877 url = f'https://www.youtube.com/shorts/{video_id}'
878
879 return {
880 '_type': 'url',
881 'ie_key': YoutubeIE.ie_key(),
882 'id': video_id,
883 'url': url,
884 'title': title,
885 'description': description,
886 'duration': duration,
887 'view_count': view_count,
888 'uploader': uploader,
889 'channel_id': channel_id,
890 'thumbnails': thumbnails,
891 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
892 if self._configuration_arg('approximate_date', ie_key='youtubetab')
893 else None),
894 'live_status': ('is_upcoming' if scheduled_timestamp is not None
895 else 'was_live' if 'streamed' in time_text.lower()
896 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
897 else None),
898 'release_timestamp': scheduled_timestamp,
899 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
900 }
901
902
903 class YoutubeIE(YoutubeBaseInfoExtractor):
904 IE_DESC = 'YouTube'
905 _VALID_URL = r"""(?x)^
906 (
907 (?:https?://|//) # http(s):// or protocol-independent URL
908 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
909 (?:www\.)?deturl\.com/www\.youtube\.com|
910 (?:www\.)?pwnyoutube\.com|
911 (?:www\.)?hooktube\.com|
912 (?:www\.)?yourepeat\.com|
913 tube\.majestyc\.net|
914 %(invidious)s|
915 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
916 (?:.*?\#/)? # handle anchor (#/) redirect urls
917 (?: # the various things that can precede the ID:
918 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
919 |(?: # or the v= param in all its forms
920 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
921 (?:\?|\#!?) # the params delimiter ? or # or #!
922 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
923 v=
924 )
925 ))
926 |(?:
927 youtu\.be| # just youtu.be/xxxx
928 vid\.plus| # or vid.plus/xxxx
929 zwearz\.com/watch| # or zwearz.com/watch/xxxx
930 %(invidious)s
931 )/
932 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
933 )
934 )? # all until now is optional -> you can pass the naked ID
935 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
936 (?(1).+)? # if we found the ID, everything can follow
937 (?:\#|$)""" % {
938 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
939 }
940 _PLAYER_INFO_RE = (
941 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
942 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$',
943 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
944 )
945 _formats = {
946 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
947 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
948 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
949 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
950 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
951 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
952 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
953 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
954 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
955 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
956 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
957 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
958 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
959 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
960 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
961 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
962 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
963 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
964
965
966 # 3D videos
967 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
968 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
969 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
970 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
971 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
972 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
973 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
974
975 # Apple HTTP Live Streaming
976 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
977 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
978 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
979 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
980 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
981 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
982 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
983 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
984
985 # DASH mp4 video
986 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
987 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
988 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
989 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
990 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
991 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
992 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
993 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
994 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
995 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
996 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
997 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
998
999 # Dash mp4 audio
1000 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1001 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1002 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1003 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1004 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1005 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1006 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1007
1008 # Dash webm
1009 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1010 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1011 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1012 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1013 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1014 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1015 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1016 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1017 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1018 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1019 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1020 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1021 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1022 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1023 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1024 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1025 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1026 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1027 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1028 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1029 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1030 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1031
1032 # Dash webm audio
1033 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1034 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1035
1036 # Dash webm audio with opus inside
1037 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1038 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1039 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1040
1041 # RTMP (unnamed)
1042 '_rtmp': {'protocol': 'rtmp'},
1043
1044 # av01 video only formats sometimes served with "unknown" codecs
1045 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1046 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1047 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1048 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1049 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1050 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1051 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1052 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1053 }
1054 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1055
1056 _GEO_BYPASS = False
1057
1058 IE_NAME = 'youtube'
1059 _TESTS = [
1060 {
1061 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
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 'start_time': 1,
1085 'end_time': 9,
1086 'channel_follower_count': int
1087 }
1088 },
1089 {
1090 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1091 'note': 'Embed-only video (#1746)',
1092 'info_dict': {
1093 'id': 'yZIXLfi8CZQ',
1094 'ext': 'mp4',
1095 'upload_date': '20120608',
1096 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1097 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1098 'uploader': 'SET India',
1099 'uploader_id': 'setindia',
1100 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1101 'age_limit': 18,
1102 },
1103 'skip': 'Private video',
1104 },
1105 {
1106 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1107 'note': 'Use the first video ID in the URL',
1108 'info_dict': {
1109 'id': 'BaW_jenozKc',
1110 'ext': 'mp4',
1111 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1112 'uploader': 'Philipp Hagemeister',
1113 'uploader_id': 'phihag',
1114 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1115 'channel': 'Philipp Hagemeister',
1116 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1117 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1118 'upload_date': '20121002',
1119 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1120 'categories': ['Science & Technology'],
1121 'tags': ['youtube-dl'],
1122 'duration': 10,
1123 'view_count': int,
1124 'like_count': int,
1125 'availability': 'public',
1126 'playable_in_embed': True,
1127 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1128 'live_status': 'not_live',
1129 'age_limit': 0,
1130 'channel_follower_count': int
1131 },
1132 'params': {
1133 'skip_download': True,
1134 },
1135 },
1136 {
1137 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1138 'note': '256k DASH audio (format 141) via DASH manifest',
1139 'info_dict': {
1140 'id': 'a9LDPn-MO4I',
1141 'ext': 'm4a',
1142 'upload_date': '20121002',
1143 'uploader_id': '8KVIDEO',
1144 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1145 'description': '',
1146 'uploader': '8KVIDEO',
1147 'title': 'UHDTV TEST 8K VIDEO.mp4'
1148 },
1149 'params': {
1150 'youtube_include_dash_manifest': True,
1151 'format': '141',
1152 },
1153 'skip': 'format 141 not served anymore',
1154 },
1155 # DASH manifest with encrypted signature
1156 {
1157 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1158 'info_dict': {
1159 'id': 'IB3lcPjvWLA',
1160 'ext': 'm4a',
1161 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1162 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1163 'duration': 244,
1164 'uploader': 'AfrojackVEVO',
1165 'uploader_id': 'AfrojackVEVO',
1166 'upload_date': '20131011',
1167 'abr': 129.495,
1168 'like_count': int,
1169 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1170 'playable_in_embed': True,
1171 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1172 'view_count': int,
1173 'track': 'The Spark',
1174 'live_status': 'not_live',
1175 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1176 'channel': 'Afrojack',
1177 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1178 'tags': 'count:19',
1179 'availability': 'public',
1180 'categories': ['Music'],
1181 'age_limit': 0,
1182 'alt_title': 'The Spark',
1183 'channel_follower_count': int
1184 },
1185 'params': {
1186 'youtube_include_dash_manifest': True,
1187 'format': '141/bestaudio[ext=m4a]',
1188 },
1189 },
1190 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1191 {
1192 'note': 'Embed allowed age-gate video',
1193 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1194 'info_dict': {
1195 'id': 'HtVdAasjOgU',
1196 'ext': 'mp4',
1197 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1198 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1199 'duration': 142,
1200 'uploader': 'The Witcher',
1201 'uploader_id': 'WitcherGame',
1202 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1203 'upload_date': '20140605',
1204 'age_limit': 18,
1205 'categories': ['Gaming'],
1206 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1207 'availability': 'needs_auth',
1208 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1209 'like_count': int,
1210 'channel': 'The Witcher',
1211 'live_status': 'not_live',
1212 'tags': 'count:17',
1213 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1214 'playable_in_embed': True,
1215 'view_count': int,
1216 'channel_follower_count': int
1217 },
1218 },
1219 {
1220 'note': 'Age-gate video with embed allowed in public site',
1221 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1222 'info_dict': {
1223 'id': 'HsUATh_Nc2U',
1224 'ext': 'mp4',
1225 'title': 'Godzilla 2 (Official Video)',
1226 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1227 'upload_date': '20200408',
1228 'uploader_id': 'FlyingKitty900',
1229 'uploader': 'FlyingKitty',
1230 'age_limit': 18,
1231 'availability': 'needs_auth',
1232 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1233 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1234 'channel': 'FlyingKitty',
1235 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1236 'view_count': int,
1237 'categories': ['Entertainment'],
1238 'live_status': 'not_live',
1239 'tags': ['Flyingkitty', 'godzilla 2'],
1240 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1241 'like_count': int,
1242 'duration': 177,
1243 'playable_in_embed': True,
1244 'channel_follower_count': int
1245 },
1246 },
1247 {
1248 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1249 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1250 'info_dict': {
1251 'id': 'Tq92D6wQ1mg',
1252 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1253 'ext': 'mp4',
1254 'upload_date': '20191228',
1255 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1256 'uploader': 'Projekt Melody',
1257 'description': 'md5:17eccca93a786d51bc67646756894066',
1258 'age_limit': 18,
1259 'like_count': int,
1260 'availability': 'needs_auth',
1261 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1262 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1263 'view_count': int,
1264 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1265 'channel': 'Projekt Melody',
1266 'live_status': 'not_live',
1267 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1268 'playable_in_embed': True,
1269 'categories': ['Entertainment'],
1270 'duration': 106,
1271 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1272 'channel_follower_count': int
1273 },
1274 },
1275 {
1276 'note': 'Non-Agegated non-embeddable video',
1277 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1278 'info_dict': {
1279 'id': 'MeJVWBSsPAY',
1280 'ext': 'mp4',
1281 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1282 'uploader': 'Herr Lurik',
1283 'uploader_id': 'st3in234',
1284 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1285 'upload_date': '20130730',
1286 'track': 'Such mich find mich',
1287 'age_limit': 0,
1288 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1289 'like_count': int,
1290 'playable_in_embed': False,
1291 'creator': 'OOMPH!',
1292 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1293 'view_count': int,
1294 'alt_title': 'Such mich find mich',
1295 'duration': 210,
1296 'channel': 'Herr Lurik',
1297 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1298 'categories': ['Music'],
1299 'availability': 'public',
1300 'uploader_url': 'http://www.youtube.com/user/st3in234',
1301 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1302 'live_status': 'not_live',
1303 'artist': 'OOMPH!',
1304 'channel_follower_count': int
1305 },
1306 },
1307 {
1308 'note': 'Non-bypassable age-gated video',
1309 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1310 'only_matching': True,
1311 },
1312 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1313 # YouTube Red ad is not captured for creator
1314 {
1315 'url': '__2ABJjxzNo',
1316 'info_dict': {
1317 'id': '__2ABJjxzNo',
1318 'ext': 'mp4',
1319 'duration': 266,
1320 'upload_date': '20100430',
1321 'uploader_id': 'deadmau5',
1322 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1323 'creator': 'deadmau5',
1324 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1325 'uploader': 'deadmau5',
1326 'title': 'Deadmau5 - Some Chords (HD)',
1327 'alt_title': 'Some Chords',
1328 'availability': 'public',
1329 'tags': 'count:14',
1330 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1331 'view_count': int,
1332 'live_status': 'not_live',
1333 'channel': 'deadmau5',
1334 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1335 'like_count': int,
1336 'track': 'Some Chords',
1337 'artist': 'deadmau5',
1338 'playable_in_embed': True,
1339 'age_limit': 0,
1340 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1341 'categories': ['Music'],
1342 'album': 'Some Chords',
1343 'channel_follower_count': int
1344 },
1345 'expected_warnings': [
1346 'DASH manifest missing',
1347 ]
1348 },
1349 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1350 {
1351 'url': 'lqQg6PlCWgI',
1352 'info_dict': {
1353 'id': 'lqQg6PlCWgI',
1354 'ext': 'mp4',
1355 'duration': 6085,
1356 'upload_date': '20150827',
1357 'uploader_id': 'olympic',
1358 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1359 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1360 'uploader': 'Olympics',
1361 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1362 'like_count': int,
1363 'release_timestamp': 1343767800,
1364 'playable_in_embed': True,
1365 'categories': ['Sports'],
1366 'release_date': '20120731',
1367 'channel': 'Olympics',
1368 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1369 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1370 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1371 'age_limit': 0,
1372 'availability': 'public',
1373 'live_status': 'was_live',
1374 'view_count': int,
1375 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1376 'channel_follower_count': int
1377 },
1378 'params': {
1379 'skip_download': 'requires avconv',
1380 }
1381 },
1382 # Non-square pixels
1383 {
1384 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1385 'info_dict': {
1386 'id': '_b-2C3KPAM0',
1387 'ext': 'mp4',
1388 'stretched_ratio': 16 / 9.,
1389 'duration': 85,
1390 'upload_date': '20110310',
1391 'uploader_id': 'AllenMeow',
1392 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1393 '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',
1394 'uploader': '孫ᄋᄅ',
1395 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1396 'playable_in_embed': True,
1397 'channel': '孫ᄋᄅ',
1398 'age_limit': 0,
1399 'tags': 'count:11',
1400 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1401 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1402 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1403 'view_count': int,
1404 'categories': ['People & Blogs'],
1405 'like_count': int,
1406 'live_status': 'not_live',
1407 'availability': 'unlisted',
1408 'channel_follower_count': int
1409 },
1410 },
1411 # url_encoded_fmt_stream_map is empty string
1412 {
1413 'url': 'qEJwOuvDf7I',
1414 'info_dict': {
1415 'id': 'qEJwOuvDf7I',
1416 'ext': 'webm',
1417 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1418 'description': '',
1419 'upload_date': '20150404',
1420 'uploader_id': 'spbelect',
1421 'uploader': 'Наблюдатели Петербурга',
1422 },
1423 'params': {
1424 'skip_download': 'requires avconv',
1425 },
1426 'skip': 'This live event has ended.',
1427 },
1428 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1429 {
1430 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1431 'info_dict': {
1432 'id': 'FIl7x6_3R5Y',
1433 'ext': 'webm',
1434 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1435 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1436 'duration': 220,
1437 'upload_date': '20150625',
1438 'uploader_id': 'dorappi2000',
1439 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1440 'uploader': 'dorappi2000',
1441 'formats': 'mincount:31',
1442 },
1443 'skip': 'not actual anymore',
1444 },
1445 # DASH manifest with segment_list
1446 {
1447 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1448 'md5': '8ce563a1d667b599d21064e982ab9e31',
1449 'info_dict': {
1450 'id': 'CsmdDsKjzN8',
1451 'ext': 'mp4',
1452 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1453 'uploader': 'Airtek',
1454 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1455 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1456 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1457 },
1458 'params': {
1459 'youtube_include_dash_manifest': True,
1460 'format': '135', # bestvideo
1461 },
1462 'skip': 'This live event has ended.',
1463 },
1464 {
1465 # Multifeed videos (multiple cameras), URL is for Main Camera
1466 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1467 'info_dict': {
1468 'id': 'jvGDaLqkpTg',
1469 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1470 'description': 'md5:e03b909557865076822aa169218d6a5d',
1471 },
1472 'playlist': [{
1473 'info_dict': {
1474 'id': 'jvGDaLqkpTg',
1475 'ext': 'mp4',
1476 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1477 'description': 'md5:e03b909557865076822aa169218d6a5d',
1478 'duration': 10643,
1479 'upload_date': '20161111',
1480 'uploader': 'Team PGP',
1481 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1482 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1483 },
1484 }, {
1485 'info_dict': {
1486 'id': '3AKt1R1aDnw',
1487 'ext': 'mp4',
1488 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1489 'description': 'md5:e03b909557865076822aa169218d6a5d',
1490 'duration': 10991,
1491 'upload_date': '20161111',
1492 'uploader': 'Team PGP',
1493 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1494 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1495 },
1496 }, {
1497 'info_dict': {
1498 'id': 'RtAMM00gpVc',
1499 'ext': 'mp4',
1500 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1501 'description': 'md5:e03b909557865076822aa169218d6a5d',
1502 'duration': 10995,
1503 'upload_date': '20161111',
1504 'uploader': 'Team PGP',
1505 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1506 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1507 },
1508 }, {
1509 'info_dict': {
1510 'id': '6N2fdlP3C5U',
1511 'ext': 'mp4',
1512 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1513 'description': 'md5:e03b909557865076822aa169218d6a5d',
1514 'duration': 10990,
1515 'upload_date': '20161111',
1516 'uploader': 'Team PGP',
1517 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1518 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1519 },
1520 }],
1521 'params': {
1522 'skip_download': True,
1523 },
1524 'skip': 'Not multifeed anymore',
1525 },
1526 {
1527 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1528 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1529 'info_dict': {
1530 'id': 'gVfLd0zydlo',
1531 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1532 },
1533 'playlist_count': 2,
1534 'skip': 'Not multifeed anymore',
1535 },
1536 {
1537 'url': 'https://vid.plus/FlRa-iH7PGw',
1538 'only_matching': True,
1539 },
1540 {
1541 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1542 'only_matching': True,
1543 },
1544 {
1545 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1546 # Also tests cut-off URL expansion in video description (see
1547 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1548 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1549 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1550 'info_dict': {
1551 'id': 'lsguqyKfVQg',
1552 'ext': 'mp4',
1553 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1554 'alt_title': 'Dark Walk',
1555 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1556 'duration': 133,
1557 'upload_date': '20151119',
1558 'uploader_id': 'IronSoulElf',
1559 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1560 'uploader': 'IronSoulElf',
1561 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1562 'track': 'Dark Walk',
1563 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1564 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1565 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1566 'categories': ['Film & Animation'],
1567 'view_count': int,
1568 'live_status': 'not_live',
1569 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1570 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1571 'tags': 'count:13',
1572 'availability': 'public',
1573 'channel': 'IronSoulElf',
1574 'playable_in_embed': True,
1575 'like_count': int,
1576 'age_limit': 0,
1577 'channel_follower_count': int
1578 },
1579 'params': {
1580 'skip_download': True,
1581 },
1582 },
1583 {
1584 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1585 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1586 'only_matching': True,
1587 },
1588 {
1589 # Video with yt:stretch=17:0
1590 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1591 'info_dict': {
1592 'id': 'Q39EVAstoRM',
1593 'ext': 'mp4',
1594 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1595 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1596 'upload_date': '20151107',
1597 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1598 'uploader': 'CH GAMER DROID',
1599 },
1600 'params': {
1601 'skip_download': True,
1602 },
1603 'skip': 'This video does not exist.',
1604 },
1605 {
1606 # Video with incomplete 'yt:stretch=16:'
1607 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1608 'only_matching': True,
1609 },
1610 {
1611 # Video licensed under Creative Commons
1612 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1613 'info_dict': {
1614 'id': 'M4gD1WSo5mA',
1615 'ext': 'mp4',
1616 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1617 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1618 'duration': 721,
1619 'upload_date': '20150128',
1620 'uploader_id': 'BerkmanCenter',
1621 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1622 'uploader': 'The Berkman Klein Center for Internet & Society',
1623 'license': 'Creative Commons Attribution license (reuse allowed)',
1624 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1625 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1626 'like_count': int,
1627 'age_limit': 0,
1628 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1629 'channel': 'The Berkman Klein Center for Internet & Society',
1630 'availability': 'public',
1631 'view_count': int,
1632 'categories': ['Education'],
1633 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1634 'live_status': 'not_live',
1635 'playable_in_embed': True,
1636 'channel_follower_count': int
1637 },
1638 'params': {
1639 'skip_download': True,
1640 },
1641 },
1642 {
1643 # Channel-like uploader_url
1644 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1645 'info_dict': {
1646 'id': 'eQcmzGIKrzg',
1647 'ext': 'mp4',
1648 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1649 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1650 'duration': 4060,
1651 'upload_date': '20151120',
1652 'uploader': 'Bernie Sanders',
1653 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1654 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1655 'license': 'Creative Commons Attribution license (reuse allowed)',
1656 'playable_in_embed': True,
1657 'tags': 'count:12',
1658 'like_count': int,
1659 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1660 'age_limit': 0,
1661 'availability': 'public',
1662 'categories': ['News & Politics'],
1663 'channel': 'Bernie Sanders',
1664 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1665 'view_count': int,
1666 'live_status': 'not_live',
1667 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1668 'channel_follower_count': int
1669 },
1670 'params': {
1671 'skip_download': True,
1672 },
1673 },
1674 {
1675 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1676 'only_matching': True,
1677 },
1678 {
1679 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1680 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1681 'only_matching': True,
1682 },
1683 {
1684 # Rental video preview
1685 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1686 'info_dict': {
1687 'id': 'uGpuVWrhIzE',
1688 'ext': 'mp4',
1689 'title': 'Piku - Trailer',
1690 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1691 'upload_date': '20150811',
1692 'uploader': 'FlixMatrix',
1693 'uploader_id': 'FlixMatrixKaravan',
1694 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1695 'license': 'Standard YouTube License',
1696 },
1697 'params': {
1698 'skip_download': True,
1699 },
1700 'skip': 'This video is not available.',
1701 },
1702 {
1703 # YouTube Red video with episode data
1704 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1705 'info_dict': {
1706 'id': 'iqKdEhx-dD4',
1707 'ext': 'mp4',
1708 'title': 'Isolation - Mind Field (Ep 1)',
1709 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1710 'duration': 2085,
1711 'upload_date': '20170118',
1712 'uploader': 'Vsauce',
1713 'uploader_id': 'Vsauce',
1714 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1715 'series': 'Mind Field',
1716 'season_number': 1,
1717 'episode_number': 1,
1718 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1719 'tags': 'count:12',
1720 'view_count': int,
1721 'availability': 'public',
1722 'age_limit': 0,
1723 'channel': 'Vsauce',
1724 'episode': 'Episode 1',
1725 'categories': ['Entertainment'],
1726 'season': 'Season 1',
1727 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1728 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1729 'like_count': int,
1730 'playable_in_embed': True,
1731 'live_status': 'not_live',
1732 'channel_follower_count': int
1733 },
1734 'params': {
1735 'skip_download': True,
1736 },
1737 'expected_warnings': [
1738 'Skipping DASH manifest',
1739 ],
1740 },
1741 {
1742 # The following content has been identified by the YouTube community
1743 # as inappropriate or offensive to some audiences.
1744 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1745 'info_dict': {
1746 'id': '6SJNVb0GnPI',
1747 'ext': 'mp4',
1748 'title': 'Race Differences in Intelligence',
1749 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1750 'duration': 965,
1751 'upload_date': '20140124',
1752 'uploader': 'New Century Foundation',
1753 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1754 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1755 },
1756 'params': {
1757 'skip_download': True,
1758 },
1759 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1760 },
1761 {
1762 # itag 212
1763 'url': '1t24XAntNCY',
1764 'only_matching': True,
1765 },
1766 {
1767 # geo restricted to JP
1768 'url': 'sJL6WA-aGkQ',
1769 'only_matching': True,
1770 },
1771 {
1772 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1773 'only_matching': True,
1774 },
1775 {
1776 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1777 'only_matching': True,
1778 },
1779 {
1780 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1781 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1782 'only_matching': True,
1783 },
1784 {
1785 # DRM protected
1786 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1787 'only_matching': True,
1788 },
1789 {
1790 # Video with unsupported adaptive stream type formats
1791 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1792 'info_dict': {
1793 'id': 'Z4Vy8R84T1U',
1794 'ext': 'mp4',
1795 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1796 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1797 'duration': 433,
1798 'upload_date': '20130923',
1799 'uploader': 'Amelia Putri Harwita',
1800 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1801 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1802 'formats': 'maxcount:10',
1803 },
1804 'params': {
1805 'skip_download': True,
1806 'youtube_include_dash_manifest': False,
1807 },
1808 'skip': 'not actual anymore',
1809 },
1810 {
1811 # Youtube Music Auto-generated description
1812 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1813 'info_dict': {
1814 'id': 'MgNrAu2pzNs',
1815 'ext': 'mp4',
1816 'title': 'Voyeur Girl',
1817 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1818 'upload_date': '20190312',
1819 'uploader': 'Stephen - Topic',
1820 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1821 'artist': 'Stephen',
1822 'track': 'Voyeur Girl',
1823 'album': 'it\'s too much love to know my dear',
1824 'release_date': '20190313',
1825 'release_year': 2019,
1826 'alt_title': 'Voyeur Girl',
1827 'view_count': int,
1828 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1829 'playable_in_embed': True,
1830 'like_count': int,
1831 'categories': ['Music'],
1832 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1833 'channel': 'Stephen',
1834 'availability': 'public',
1835 'creator': 'Stephen',
1836 'duration': 169,
1837 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1838 'age_limit': 0,
1839 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1840 'tags': 'count:11',
1841 'live_status': 'not_live',
1842 'channel_follower_count': int
1843 },
1844 'params': {
1845 'skip_download': True,
1846 },
1847 },
1848 {
1849 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1850 'only_matching': True,
1851 },
1852 {
1853 # invalid -> valid video id redirection
1854 'url': 'DJztXj2GPfl',
1855 'info_dict': {
1856 'id': 'DJztXj2GPfk',
1857 'ext': 'mp4',
1858 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1859 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1860 'upload_date': '20090125',
1861 'uploader': 'Prochorowka',
1862 'uploader_id': 'Prochorowka',
1863 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1864 'artist': 'Panjabi MC',
1865 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1866 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1867 },
1868 'params': {
1869 'skip_download': True,
1870 },
1871 'skip': 'Video unavailable',
1872 },
1873 {
1874 # empty description results in an empty string
1875 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1876 'info_dict': {
1877 'id': 'x41yOUIvK2k',
1878 'ext': 'mp4',
1879 'title': 'IMG 3456',
1880 'description': '',
1881 'upload_date': '20170613',
1882 'uploader_id': 'ElevageOrVert',
1883 'uploader': 'ElevageOrVert',
1884 'view_count': int,
1885 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1886 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1887 'like_count': int,
1888 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1889 'tags': [],
1890 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1891 'availability': 'public',
1892 'age_limit': 0,
1893 'categories': ['Pets & Animals'],
1894 'duration': 7,
1895 'playable_in_embed': True,
1896 'live_status': 'not_live',
1897 'channel': 'ElevageOrVert',
1898 'channel_follower_count': int
1899 },
1900 'params': {
1901 'skip_download': True,
1902 },
1903 },
1904 {
1905 # with '};' inside yt initial data (see [1])
1906 # see [2] for an example with '};' inside ytInitialPlayerResponse
1907 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1908 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1909 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1910 'info_dict': {
1911 'id': 'CHqg6qOn4no',
1912 'ext': 'mp4',
1913 'title': 'Part 77 Sort a list of simple types in c#',
1914 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1915 'upload_date': '20130831',
1916 'uploader_id': 'kudvenkat',
1917 'uploader': 'kudvenkat',
1918 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1919 'like_count': int,
1920 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1921 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1922 'live_status': 'not_live',
1923 'categories': ['Education'],
1924 'availability': 'public',
1925 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1926 'tags': 'count:12',
1927 'playable_in_embed': True,
1928 'age_limit': 0,
1929 'view_count': int,
1930 'duration': 522,
1931 'channel': 'kudvenkat',
1932 'channel_follower_count': int
1933 },
1934 'params': {
1935 'skip_download': True,
1936 },
1937 },
1938 {
1939 # another example of '};' in ytInitialData
1940 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1941 'only_matching': True,
1942 },
1943 {
1944 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1945 'only_matching': True,
1946 },
1947 {
1948 # https://github.com/ytdl-org/youtube-dl/pull/28094
1949 'url': 'OtqTfy26tG0',
1950 'info_dict': {
1951 'id': 'OtqTfy26tG0',
1952 'ext': 'mp4',
1953 'title': 'Burn Out',
1954 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1955 'upload_date': '20141120',
1956 'uploader': 'The Cinematic Orchestra - Topic',
1957 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1958 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1959 'artist': 'The Cinematic Orchestra',
1960 'track': 'Burn Out',
1961 'album': 'Every Day',
1962 'like_count': int,
1963 'live_status': 'not_live',
1964 'alt_title': 'Burn Out',
1965 'duration': 614,
1966 'age_limit': 0,
1967 'view_count': int,
1968 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1969 'creator': 'The Cinematic Orchestra',
1970 'channel': 'The Cinematic Orchestra',
1971 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1972 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1973 'availability': 'public',
1974 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1975 'categories': ['Music'],
1976 'playable_in_embed': True,
1977 'channel_follower_count': int
1978 },
1979 'params': {
1980 'skip_download': True,
1981 },
1982 },
1983 {
1984 # controversial video, only works with bpctr when authenticated with cookies
1985 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1986 'only_matching': True,
1987 },
1988 {
1989 # controversial video, requires bpctr/contentCheckOk
1990 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1991 'info_dict': {
1992 'id': 'SZJvDhaSDnc',
1993 'ext': 'mp4',
1994 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1995 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1996 'uploader': 'CBS Mornings',
1997 'uploader_id': 'CBSThisMorning',
1998 'upload_date': '20140716',
1999 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2000 'duration': 170,
2001 'categories': ['News & Politics'],
2002 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2003 'view_count': int,
2004 'channel': 'CBS Mornings',
2005 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2006 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2007 'age_limit': 18,
2008 'availability': 'needs_auth',
2009 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2010 'like_count': int,
2011 'live_status': 'not_live',
2012 'playable_in_embed': True,
2013 'channel_follower_count': int
2014 }
2015 },
2016 {
2017 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2018 'url': 'cBvYw8_A0vQ',
2019 'info_dict': {
2020 'id': 'cBvYw8_A0vQ',
2021 'ext': 'mp4',
2022 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2023 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2024 'upload_date': '20201120',
2025 'uploader': 'Walk around Japan',
2026 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2027 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2028 'duration': 1456,
2029 'categories': ['Travel & Events'],
2030 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2031 'view_count': int,
2032 'channel': 'Walk around Japan',
2033 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2034 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2035 'age_limit': 0,
2036 'availability': 'public',
2037 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2038 'live_status': 'not_live',
2039 'playable_in_embed': True,
2040 'channel_follower_count': int
2041 },
2042 'params': {
2043 'skip_download': True,
2044 },
2045 }, {
2046 # Has multiple audio streams
2047 'url': 'WaOKSUlf4TM',
2048 'only_matching': True
2049 }, {
2050 # Requires Premium: has format 141 when requested using YTM url
2051 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2052 'only_matching': True
2053 }, {
2054 # multiple subtitles with same lang_code
2055 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2056 'only_matching': True,
2057 }, {
2058 # Force use android client fallback
2059 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2060 'info_dict': {
2061 'id': 'YOelRv7fMxY',
2062 'title': 'DIGGING A SECRET TUNNEL Part 1',
2063 'ext': '3gp',
2064 'upload_date': '20210624',
2065 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2066 'uploader': 'colinfurze',
2067 'uploader_id': 'colinfurze',
2068 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2069 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2070 'duration': 596,
2071 'categories': ['Entertainment'],
2072 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2073 'view_count': int,
2074 'channel': 'colinfurze',
2075 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2076 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2077 'age_limit': 0,
2078 'availability': 'public',
2079 'like_count': int,
2080 'live_status': 'not_live',
2081 'playable_in_embed': True,
2082 'channel_follower_count': int
2083 },
2084 'params': {
2085 'format': '17', # 3gp format available on android
2086 'extractor_args': {'youtube': {'player_client': ['android']}},
2087 },
2088 },
2089 {
2090 # Skip download of additional client configs (remix client config in this case)
2091 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2092 'only_matching': True,
2093 'params': {
2094 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2095 },
2096 }, {
2097 # shorts
2098 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2099 'only_matching': True,
2100 }, {
2101 'note': 'Storyboards',
2102 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2103 'info_dict': {
2104 'id': '5KLPxDtMqe8',
2105 'ext': 'mhtml',
2106 'format_id': 'sb0',
2107 'title': 'Your Brain is Plastic',
2108 'uploader_id': 'scishow',
2109 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2110 'upload_date': '20140324',
2111 'uploader': 'SciShow',
2112 'like_count': int,
2113 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2114 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2115 'view_count': int,
2116 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2117 'playable_in_embed': True,
2118 'tags': 'count:12',
2119 'uploader_url': 'http://www.youtube.com/user/scishow',
2120 'availability': 'public',
2121 'channel': 'SciShow',
2122 'live_status': 'not_live',
2123 'duration': 248,
2124 'categories': ['Education'],
2125 'age_limit': 0,
2126 'channel_follower_count': int
2127 }, 'params': {'format': 'mhtml', 'skip_download': True}
2128 }, {
2129 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2130 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2131 'info_dict': {
2132 'id': '2NUZ8W2llS4',
2133 'ext': 'mp4',
2134 'title': 'The NP that test your phone performance 🙂',
2135 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2136 'uploader': 'Leon Nguyen',
2137 'uploader_id': 'VNSXIII',
2138 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2139 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2140 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2141 'duration': 21,
2142 'view_count': int,
2143 'age_limit': 0,
2144 'categories': ['Gaming'],
2145 'tags': 'count:23',
2146 'playable_in_embed': True,
2147 'live_status': 'not_live',
2148 'upload_date': '20220103',
2149 'like_count': int,
2150 'availability': 'public',
2151 'channel': 'Leon Nguyen',
2152 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2153 'channel_follower_count': int
2154 }
2155 }, {
2156 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2157 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2158 'info_dict': {
2159 'id': 'mzZzzBU6lrM',
2160 'ext': 'mp4',
2161 'title': 'I Met GeorgeNotFound In Real Life...',
2162 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2163 'uploader': 'Quackity',
2164 'uploader_id': 'QuackityHQ',
2165 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2166 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2167 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2168 'duration': 955,
2169 'view_count': int,
2170 'age_limit': 0,
2171 'categories': ['Entertainment'],
2172 'tags': 'count:26',
2173 'playable_in_embed': True,
2174 'live_status': 'not_live',
2175 'release_timestamp': 1641172509,
2176 'release_date': '20220103',
2177 'upload_date': '20220103',
2178 'like_count': int,
2179 'availability': 'public',
2180 'channel': 'Quackity',
2181 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2182 'channel_follower_count': int
2183 }
2184 },
2185 { # continuous livestream. Microformat upload date should be preferred.
2186 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2187 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2188 'info_dict': {
2189 'id': 'kgx4WGK0oNU',
2190 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2191 'ext': 'mp4',
2192 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2193 'availability': 'public',
2194 'age_limit': 0,
2195 'release_timestamp': 1637975704,
2196 'upload_date': '20210619',
2197 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2198 'live_status': 'is_live',
2199 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2200 'uploader': '阿鲍Abao',
2201 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2202 'channel': 'Abao in Tokyo',
2203 'channel_follower_count': int,
2204 'release_date': '20211127',
2205 'tags': 'count:39',
2206 'categories': ['People & Blogs'],
2207 'like_count': int,
2208 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2209 'view_count': int,
2210 'playable_in_embed': True,
2211 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2212 },
2213 'params': {'skip_download': True}
2214 }, {
2215 # Story. Requires specific player params to work.
2216 # Note: stories get removed after some period of time
2217 'url': 'https://www.youtube.com/watch?v=yN3x1t3sieA',
2218 'info_dict': {
2219 'id': 'yN3x1t3sieA',
2220 'ext': 'mp4',
2221 'uploader': 'Linus Tech Tips',
2222 'duration': 13,
2223 'channel': 'Linus Tech Tips',
2224 'playable_in_embed': True,
2225 'tags': [],
2226 'age_limit': 0,
2227 'uploader_url': 'http://www.youtube.com/user/LinusTechTips',
2228 'upload_date': '20220402',
2229 'thumbnail': 'https://i.ytimg.com/vi_webp/yN3x1t3sieA/maxresdefault.webp',
2230 'title': 'Story',
2231 'live_status': 'not_live',
2232 'uploader_id': 'LinusTechTips',
2233 'view_count': int,
2234 'description': '',
2235 'channel_id': 'UCXuqSBlHAE6Xw-yeJA0Tunw',
2236 'categories': ['Science & Technology'],
2237 'channel_url': 'https://www.youtube.com/channel/UCXuqSBlHAE6Xw-yeJA0Tunw',
2238 'availability': 'unlisted',
2239 }
2240 }
2241 ]
2242
2243 @classmethod
2244 def suitable(cls, url):
2245 from ..utils import parse_qs
2246
2247 qs = parse_qs(url)
2248 if qs.get('list', [None])[0]:
2249 return False
2250 return super().suitable(url)
2251
2252 def __init__(self, *args, **kwargs):
2253 super().__init__(*args, **kwargs)
2254 self._code_cache = {}
2255 self._player_cache = {}
2256
2257 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2258 lock = threading.Lock()
2259
2260 is_live = True
2261 start_time = time.time()
2262 formats = [f for f in formats if f.get('is_from_start')]
2263
2264 def refetch_manifest(format_id, delay):
2265 nonlocal formats, start_time, is_live
2266 if time.time() <= start_time + delay:
2267 return
2268
2269 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2270 video_details = traverse_obj(
2271 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2272 microformats = traverse_obj(
2273 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2274 expected_type=dict, default=[])
2275 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2276 start_time = time.time()
2277
2278 def mpd_feed(format_id, delay):
2279 """
2280 @returns (manifest_url, manifest_stream_number, is_live) or None
2281 """
2282 with lock:
2283 refetch_manifest(format_id, delay)
2284
2285 f = next((f for f in formats if f['format_id'] == format_id), None)
2286 if not f:
2287 if not is_live:
2288 self.to_screen(f'{video_id}: Video is no longer live')
2289 else:
2290 self.report_warning(
2291 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2292 return None
2293 return f['manifest_url'], f['manifest_stream_number'], is_live
2294
2295 for f in formats:
2296 f['is_live'] = True
2297 f['protocol'] = 'http_dash_segments_generator'
2298 f['fragments'] = functools.partial(
2299 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2300
2301 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2302 FETCH_SPAN, MAX_DURATION = 5, 432000
2303
2304 mpd_url, stream_number, is_live = None, None, True
2305
2306 begin_index = 0
2307 download_start_time = ctx.get('start') or time.time()
2308
2309 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2310 if lack_early_segments:
2311 self.report_warning(bug_reports_message(
2312 'Starting download from the last 120 hours of the live stream since '
2313 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2314 lack_early_segments = True
2315
2316 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2317 fragments, fragment_base_url = None, None
2318
2319 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2320 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2321 # Obtain from MPD's maximum seq value
2322 old_mpd_url = mpd_url
2323 last_error = ctx.pop('last_error', None)
2324 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2325 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2326 or (mpd_url, stream_number, False))
2327 if not refresh_sequence:
2328 if expire_fast and not is_live:
2329 return False, last_seq
2330 elif old_mpd_url == mpd_url:
2331 return True, last_seq
2332 try:
2333 fmts, _ = self._extract_mpd_formats_and_subtitles(
2334 mpd_url, None, note=False, errnote=False, fatal=False)
2335 except ExtractorError:
2336 fmts = None
2337 if not fmts:
2338 no_fragment_score += 2
2339 return False, last_seq
2340 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2341 fragments = fmt_info['fragments']
2342 fragment_base_url = fmt_info['fragment_base_url']
2343 assert fragment_base_url
2344
2345 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2346 return True, _last_seq
2347
2348 while is_live:
2349 fetch_time = time.time()
2350 if no_fragment_score > 30:
2351 return
2352 if last_segment_url:
2353 # Obtain from "X-Head-Seqnum" header value from each segment
2354 try:
2355 urlh = self._request_webpage(
2356 last_segment_url, None, note=False, errnote=False, fatal=False)
2357 except ExtractorError:
2358 urlh = None
2359 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2360 if last_seq is None:
2361 no_fragment_score += 2
2362 last_segment_url = None
2363 continue
2364 else:
2365 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2366 no_fragment_score += 2
2367 if not should_continue:
2368 continue
2369
2370 if known_idx > last_seq:
2371 last_segment_url = None
2372 continue
2373
2374 last_seq += 1
2375
2376 if begin_index < 0 and known_idx < 0:
2377 # skip from the start when it's negative value
2378 known_idx = last_seq + begin_index
2379 if lack_early_segments:
2380 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2381 try:
2382 for idx in range(known_idx, last_seq):
2383 # do not update sequence here or you'll get skipped some part of it
2384 should_continue, _ = _extract_sequence_from_mpd(False, False)
2385 if not should_continue:
2386 known_idx = idx - 1
2387 raise ExtractorError('breaking out of outer loop')
2388 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2389 yield {
2390 'url': last_segment_url,
2391 }
2392 if known_idx == last_seq:
2393 no_fragment_score += 5
2394 else:
2395 no_fragment_score = 0
2396 known_idx = last_seq
2397 except ExtractorError:
2398 continue
2399
2400 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2401
2402 def _extract_player_url(self, *ytcfgs, webpage=None):
2403 player_url = traverse_obj(
2404 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2405 get_all=False, expected_type=compat_str)
2406 if not player_url:
2407 return
2408 return urljoin('https://www.youtube.com', player_url)
2409
2410 def _download_player_url(self, video_id, fatal=False):
2411 res = self._download_webpage(
2412 'https://www.youtube.com/iframe_api',
2413 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2414 if res:
2415 player_version = self._search_regex(
2416 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2417 if player_version:
2418 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2419
2420 def _signature_cache_id(self, example_sig):
2421 """ Return a string representation of a signature """
2422 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2423
2424 @classmethod
2425 def _extract_player_info(cls, player_url):
2426 for player_re in cls._PLAYER_INFO_RE:
2427 id_m = re.search(player_re, player_url)
2428 if id_m:
2429 break
2430 else:
2431 raise ExtractorError('Cannot identify player %r' % player_url)
2432 return id_m.group('id')
2433
2434 def _load_player(self, video_id, player_url, fatal=True):
2435 player_id = self._extract_player_info(player_url)
2436 if player_id not in self._code_cache:
2437 code = self._download_webpage(
2438 player_url, video_id, fatal=fatal,
2439 note='Downloading player ' + player_id,
2440 errnote='Download of %s failed' % player_url)
2441 if code:
2442 self._code_cache[player_id] = code
2443 return self._code_cache.get(player_id)
2444
2445 def _extract_signature_function(self, video_id, player_url, example_sig):
2446 player_id = self._extract_player_info(player_url)
2447
2448 # Read from filesystem cache
2449 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2450 assert os.path.basename(func_id) == func_id
2451
2452 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2453 if cache_spec is not None:
2454 return lambda s: ''.join(s[i] for i in cache_spec)
2455
2456 code = self._load_player(video_id, player_url)
2457 if code:
2458 res = self._parse_sig_js(code)
2459
2460 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2461 cache_res = res(test_string)
2462 cache_spec = [ord(c) for c in cache_res]
2463
2464 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2465 return res
2466
2467 def _print_sig_code(self, func, example_sig):
2468 if not self.get_param('youtube_print_sig_code'):
2469 return
2470
2471 def gen_sig_code(idxs):
2472 def _genslice(start, end, step):
2473 starts = '' if start == 0 else str(start)
2474 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2475 steps = '' if step == 1 else (':%d' % step)
2476 return f's[{starts}{ends}{steps}]'
2477
2478 step = None
2479 # Quelch pyflakes warnings - start will be set when step is set
2480 start = '(Never used)'
2481 for i, prev in zip(idxs[1:], idxs[:-1]):
2482 if step is not None:
2483 if i - prev == step:
2484 continue
2485 yield _genslice(start, prev, step)
2486 step = None
2487 continue
2488 if i - prev in [-1, 1]:
2489 step = i - prev
2490 start = prev
2491 continue
2492 else:
2493 yield 's[%d]' % prev
2494 if step is None:
2495 yield 's[%d]' % i
2496 else:
2497 yield _genslice(start, i, step)
2498
2499 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2500 cache_res = func(test_string)
2501 cache_spec = [ord(c) for c in cache_res]
2502 expr_code = ' + '.join(gen_sig_code(cache_spec))
2503 signature_id_tuple = '(%s)' % (
2504 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2505 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2506 ' return %s\n') % (signature_id_tuple, expr_code)
2507 self.to_screen('Extracted signature function:\n' + code)
2508
2509 def _parse_sig_js(self, jscode):
2510 funcname = self._search_regex(
2511 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2512 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2513 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2514 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2515 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}\(a,\d+\)',
2516 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\(\s*""\s*\)',
2517 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2518 # Obsolete patterns
2519 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2520 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2521 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2522 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2523 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2524 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2525 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2526 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2527 jscode, 'Initial JS player signature function name', group='sig')
2528
2529 jsi = JSInterpreter(jscode)
2530 initial_function = jsi.extract_function(funcname)
2531 return lambda s: initial_function([s])
2532
2533 def _decrypt_signature(self, s, video_id, player_url):
2534 """Turn the encrypted s field into a working signature"""
2535
2536 if player_url is None:
2537 raise ExtractorError('Cannot decrypt signature without player_url')
2538
2539 try:
2540 player_id = (player_url, self._signature_cache_id(s))
2541 if player_id not in self._player_cache:
2542 func = self._extract_signature_function(
2543 video_id, player_url, s
2544 )
2545 self._player_cache[player_id] = func
2546 func = self._player_cache[player_id]
2547 self._print_sig_code(func, s)
2548 return func(s)
2549 except Exception as e:
2550 raise ExtractorError('Signature extraction failed: ' + traceback.format_exc(), cause=e)
2551
2552 def _decrypt_nsig(self, s, video_id, player_url):
2553 """Turn the encrypted n field into a working signature"""
2554 if player_url is None:
2555 raise ExtractorError('Cannot decrypt nsig without player_url')
2556 player_url = urljoin('https://www.youtube.com', player_url)
2557
2558 sig_id = ('nsig_value', s)
2559 if sig_id in self._player_cache:
2560 return self._player_cache[sig_id]
2561
2562 try:
2563 player_id = ('nsig', player_url)
2564 if player_id not in self._player_cache:
2565 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2566 func = self._player_cache[player_id]
2567 self._player_cache[sig_id] = func(s)
2568 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2569 return self._player_cache[sig_id]
2570 except Exception as e:
2571 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2572
2573 def _extract_n_function_name(self, jscode):
2574 nfunc, idx = self._search_regex(
2575 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2576 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2577 if not idx:
2578 return nfunc
2579 return json.loads(js_to_json(self._search_regex(
2580 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2581 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2582
2583 def _extract_n_function(self, video_id, player_url):
2584 player_id = self._extract_player_info(player_url)
2585 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2586
2587 if func_code:
2588 jsi = JSInterpreter(func_code)
2589 else:
2590 jscode = self._load_player(video_id, player_url)
2591 funcname = self._extract_n_function_name(jscode)
2592 jsi = JSInterpreter(jscode)
2593 func_code = jsi.extract_function_code(funcname)
2594 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2595
2596 if self.get_param('youtube_print_sig_code'):
2597 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2598
2599 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2600
2601 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2602 """
2603 Extract signatureTimestamp (sts)
2604 Required to tell API what sig/player version is in use.
2605 """
2606 sts = None
2607 if isinstance(ytcfg, dict):
2608 sts = int_or_none(ytcfg.get('STS'))
2609
2610 if not sts:
2611 # Attempt to extract from player
2612 if player_url is None:
2613 error_msg = 'Cannot extract signature timestamp without player_url.'
2614 if fatal:
2615 raise ExtractorError(error_msg)
2616 self.report_warning(error_msg)
2617 return
2618 code = self._load_player(video_id, player_url, fatal=fatal)
2619 if code:
2620 sts = int_or_none(self._search_regex(
2621 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2622 'JS player signature timestamp', group='sts', fatal=fatal))
2623 return sts
2624
2625 def _mark_watched(self, video_id, player_responses):
2626 playback_url = get_first(
2627 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2628 expected_type=url_or_none)
2629 if not playback_url:
2630 self.report_warning('Unable to mark watched')
2631 return
2632 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2633 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2634
2635 # cpn generation algorithm is reverse engineered from base.js.
2636 # In fact it works even with dummy cpn.
2637 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2638 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2639
2640 qs.update({
2641 'ver': ['2'],
2642 'cpn': [cpn],
2643 })
2644 playback_url = compat_urlparse.urlunparse(
2645 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2646
2647 self._download_webpage(
2648 playback_url, video_id, 'Marking watched',
2649 'Unable to mark watched', fatal=False)
2650
2651 @staticmethod
2652 def _extract_urls(webpage):
2653 # Embedded YouTube player
2654 entries = [
2655 unescapeHTML(mobj.group('url'))
2656 for mobj in re.finditer(r'''(?x)
2657 (?:
2658 <iframe[^>]+?src=|
2659 data-video-url=|
2660 <embed[^>]+?src=|
2661 embedSWF\(?:\s*|
2662 <object[^>]+data=|
2663 new\s+SWFObject\(
2664 )
2665 (["\'])
2666 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2667 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2668 \1''', webpage)]
2669
2670 # lazyYT YouTube embed
2671 entries.extend(list(map(
2672 unescapeHTML,
2673 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2674
2675 # Wordpress "YouTube Video Importer" plugin
2676 matches = re.findall(r'''(?x)<div[^>]+
2677 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2678 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2679 entries.extend(m[-1] for m in matches)
2680
2681 return entries
2682
2683 @staticmethod
2684 def _extract_url(webpage):
2685 urls = YoutubeIE._extract_urls(webpage)
2686 return urls[0] if urls else None
2687
2688 @classmethod
2689 def extract_id(cls, url):
2690 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2691 if mobj is None:
2692 raise ExtractorError('Invalid URL: %s' % url)
2693 return mobj.group('id')
2694
2695 def _extract_chapters_from_json(self, data, duration):
2696 chapter_list = traverse_obj(
2697 data, (
2698 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2699 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2700 ), expected_type=list)
2701
2702 return self._extract_chapters(
2703 chapter_list,
2704 chapter_time=lambda chapter: float_or_none(
2705 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2706 chapter_title=lambda chapter: traverse_obj(
2707 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2708 duration=duration)
2709
2710 def _extract_chapters_from_engagement_panel(self, data, duration):
2711 content_list = traverse_obj(
2712 data,
2713 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2714 expected_type=list, default=[])
2715 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2716 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2717
2718 return next((
2719 filter(None, (
2720 self._extract_chapters(
2721 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2722 chapter_time, chapter_title, duration)
2723 for contents in content_list
2724 ))), [])
2725
2726 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2727 chapters = []
2728 last_chapter = {'start_time': 0}
2729 for idx, chapter in enumerate(chapter_list or []):
2730 title = chapter_title(chapter)
2731 start_time = chapter_time(chapter)
2732 if start_time is None:
2733 continue
2734 last_chapter['end_time'] = start_time
2735 if start_time < last_chapter['start_time']:
2736 if idx == 1:
2737 chapters.pop()
2738 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2739 else:
2740 self.report_warning(f'Invalid start time for chapter "{title}"')
2741 continue
2742 last_chapter = {'start_time': start_time, 'title': title}
2743 chapters.append(last_chapter)
2744 last_chapter['end_time'] = duration
2745 return chapters
2746
2747 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2748 return self._parse_json(self._search_regex(
2749 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
2750 regex), webpage, name, default='{}'), video_id, fatal=False)
2751
2752 def _extract_comment(self, comment_renderer, parent=None):
2753 comment_id = comment_renderer.get('commentId')
2754 if not comment_id:
2755 return
2756
2757 text = self._get_text(comment_renderer, 'contentText')
2758
2759 # note: timestamp is an estimate calculated from the current time and time_text
2760 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2761 author = self._get_text(comment_renderer, 'authorText')
2762 author_id = try_get(comment_renderer,
2763 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2764
2765 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2766 lambda x: x['likeCount']), compat_str)) or 0
2767 author_thumbnail = try_get(comment_renderer,
2768 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2769
2770 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2771 is_favorited = 'creatorHeart' in (try_get(
2772 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2773 return {
2774 'id': comment_id,
2775 'text': text,
2776 'timestamp': timestamp,
2777 'time_text': time_text,
2778 'like_count': votes,
2779 'is_favorited': is_favorited,
2780 'author': author,
2781 'author_id': author_id,
2782 'author_thumbnail': author_thumbnail,
2783 'author_is_uploader': author_is_uploader,
2784 'parent': parent or 'root'
2785 }
2786
2787 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2788
2789 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2790
2791 def extract_header(contents):
2792 _continuation = None
2793 for content in contents:
2794 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2795 expected_comment_count = self._get_count(
2796 comments_header_renderer, 'countText', 'commentsCount')
2797
2798 if expected_comment_count:
2799 tracker['est_total'] = expected_comment_count
2800 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2801 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2802
2803 sort_menu_item = try_get(
2804 comments_header_renderer,
2805 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2806 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2807
2808 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2809 if not _continuation:
2810 continue
2811
2812 sort_text = str_or_none(sort_menu_item.get('title'))
2813 if not sort_text:
2814 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2815 self.to_screen('Sorting comments by %s' % sort_text.lower())
2816 break
2817 return _continuation
2818
2819 def extract_thread(contents):
2820 if not parent:
2821 tracker['current_page_thread'] = 0
2822 for content in contents:
2823 if not parent and tracker['total_parent_comments'] >= max_parents:
2824 yield
2825 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2826 comment_renderer = get_first(
2827 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2828 expected_type=dict, default={})
2829
2830 comment = self._extract_comment(comment_renderer, parent)
2831 if not comment:
2832 continue
2833
2834 tracker['running_total'] += 1
2835 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2836 yield comment
2837
2838 # Attempt to get the replies
2839 comment_replies_renderer = try_get(
2840 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2841
2842 if comment_replies_renderer:
2843 tracker['current_page_thread'] += 1
2844 comment_entries_iter = self._comment_entries(
2845 comment_replies_renderer, ytcfg, video_id,
2846 parent=comment.get('id'), tracker=tracker)
2847 yield from itertools.islice(comment_entries_iter, min(
2848 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2849
2850 # Keeps track of counts across recursive calls
2851 if not tracker:
2852 tracker = dict(
2853 running_total=0,
2854 est_total=0,
2855 current_page_thread=0,
2856 total_parent_comments=0,
2857 total_reply_comments=0)
2858
2859 # TODO: Deprecated
2860 # YouTube comments have a max depth of 2
2861 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2862 if max_depth:
2863 self._downloader.deprecation_warning(
2864 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2865 if max_depth == 1 and parent:
2866 return
2867
2868 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2869 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2870
2871 continuation = self._extract_continuation(root_continuation_data)
2872
2873 response = None
2874 is_forced_continuation = False
2875 is_first_continuation = parent is None
2876 if is_first_continuation and not continuation:
2877 # Sometimes you can get comments by generating the continuation yourself,
2878 # even if YouTube initially reports them being disabled - e.g. stories comments.
2879 # Note: if the comment section is actually disabled, YouTube may return a response with
2880 # required check_get_keys missing. So we will disable that check initially in this case.
2881 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2882 is_forced_continuation = True
2883
2884 for page_num in itertools.count(0):
2885 if not continuation:
2886 break
2887 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2888 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2889 if page_num == 0:
2890 if is_first_continuation:
2891 note_prefix = 'Downloading comment section API JSON'
2892 else:
2893 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2894 tracker['current_page_thread'], comment_prog_str)
2895 else:
2896 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2897 ' ' if parent else '', ' replies' if parent else '',
2898 page_num, comment_prog_str)
2899
2900 response = self._extract_response(
2901 item_id=None, query=continuation,
2902 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2903 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2904 is_forced_continuation = False
2905 continuation_contents = traverse_obj(
2906 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2907
2908 continuation = None
2909 for continuation_section in continuation_contents:
2910 continuation_items = traverse_obj(
2911 continuation_section,
2912 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2913 get_all=False, expected_type=list) or []
2914 if is_first_continuation:
2915 continuation = extract_header(continuation_items)
2916 is_first_continuation = False
2917 if continuation:
2918 break
2919 continue
2920
2921 for entry in extract_thread(continuation_items):
2922 if not entry:
2923 return
2924 yield entry
2925 continuation = self._extract_continuation({'contents': continuation_items})
2926 if continuation:
2927 break
2928
2929 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2930 if message and not parent and tracker['running_total'] == 0:
2931 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2932
2933 @staticmethod
2934 def _generate_comment_continuation(video_id):
2935 """
2936 Generates initial comment section continuation token from given video id
2937 """
2938 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2939 return base64.b64encode(token.encode()).decode()
2940
2941 def _get_comments(self, ytcfg, video_id, contents, webpage):
2942 """Entry for comment extraction"""
2943 def _real_comment_extract(contents):
2944 renderer = next((
2945 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2946 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2947 yield from self._comment_entries(renderer, ytcfg, video_id)
2948
2949 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2950 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2951
2952 @staticmethod
2953 def _get_checkok_params():
2954 return {'contentCheckOk': True, 'racyCheckOk': True}
2955
2956 @classmethod
2957 def _generate_player_context(cls, sts=None):
2958 context = {
2959 'html5Preference': 'HTML5_PREF_WANTS',
2960 }
2961 if sts is not None:
2962 context['signatureTimestamp'] = sts
2963 return {
2964 'playbackContext': {
2965 'contentPlaybackContext': context
2966 },
2967 **cls._get_checkok_params()
2968 }
2969
2970 @staticmethod
2971 def _is_agegated(player_response):
2972 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2973 return True
2974
2975 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2976 AGE_GATE_REASONS = (
2977 'confirm your age', 'age-restricted', 'inappropriate', # reason
2978 'age_verification_required', 'age_check_required', # status
2979 )
2980 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2981
2982 @staticmethod
2983 def _is_unplayable(player_response):
2984 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2985
2986 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2987
2988 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2989 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2990 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2991 headers = self.generate_api_headers(
2992 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
2993
2994 yt_query = {
2995 'videoId': video_id,
2996 'params': '8AEB' # enable stories
2997 }
2998 yt_query.update(self._generate_player_context(sts))
2999 return self._extract_response(
3000 item_id=video_id, ep='player', query=yt_query,
3001 ytcfg=player_ytcfg, headers=headers, fatal=True,
3002 default_client=client,
3003 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3004 ) or None
3005
3006 def _get_requested_clients(self, url, smuggled_data):
3007 requested_clients = []
3008 default = ['android', 'web']
3009 allowed_clients = sorted(
3010 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3011 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3012 for client in self._configuration_arg('player_client'):
3013 if client in allowed_clients:
3014 requested_clients.append(client)
3015 elif client == 'default':
3016 requested_clients.extend(default)
3017 elif client == 'all':
3018 requested_clients.extend(allowed_clients)
3019 else:
3020 self.report_warning(f'Skipping unsupported client {client}')
3021 if not requested_clients:
3022 requested_clients = default
3023
3024 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3025 requested_clients.extend(
3026 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3027
3028 return orderedSet(requested_clients)
3029
3030 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3031 initial_pr = None
3032 if webpage:
3033 initial_pr = self._extract_yt_initial_variable(
3034 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
3035 video_id, 'initial player response')
3036
3037 all_clients = set(clients)
3038 clients = clients[::-1]
3039 prs = []
3040
3041 def append_client(*client_names):
3042 """ Append the first client name that exists but not already used """
3043 for client_name in client_names:
3044 actual_client = _split_innertube_client(client_name)[0]
3045 if actual_client in INNERTUBE_CLIENTS:
3046 if actual_client not in all_clients:
3047 clients.append(client_name)
3048 all_clients.add(actual_client)
3049 return
3050
3051 # Android player_response does not have microFormats which are needed for
3052 # extraction of some data. So we return the initial_pr with formats
3053 # stripped out even if not requested by the user
3054 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3055 if initial_pr:
3056 pr = dict(initial_pr)
3057 pr['streamingData'] = None
3058 prs.append(pr)
3059
3060 last_error = None
3061 tried_iframe_fallback = False
3062 player_url = None
3063 while clients:
3064 client, base_client, variant = _split_innertube_client(clients.pop())
3065 player_ytcfg = master_ytcfg if client == 'web' else {}
3066 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3067 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3068
3069 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3070 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3071 if 'js' in self._configuration_arg('player_skip'):
3072 require_js_player = False
3073 player_url = None
3074
3075 if not player_url and not tried_iframe_fallback and require_js_player:
3076 player_url = self._download_player_url(video_id)
3077 tried_iframe_fallback = True
3078
3079 try:
3080 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3081 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3082 except ExtractorError as e:
3083 if last_error:
3084 self.report_warning(last_error)
3085 last_error = e
3086 continue
3087
3088 if pr:
3089 prs.append(pr)
3090
3091 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3092 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3093 append_client(f'{base_client}_creator')
3094 elif self._is_agegated(pr):
3095 if variant == 'tv_embedded':
3096 append_client(f'{base_client}_embedded')
3097 elif not variant:
3098 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3099
3100 if last_error:
3101 if not len(prs):
3102 raise last_error
3103 self.report_warning(last_error)
3104 return prs, player_url
3105
3106 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3107 itags, stream_ids = {}, []
3108 itag_qualities, res_qualities = {}, {}
3109 q = qualities([
3110 # Normally tiny is the smallest video-only formats. But
3111 # audio-only formats with unknown quality may get tagged as tiny
3112 'tiny',
3113 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3114 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3115 ])
3116 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3117
3118 for fmt in streaming_formats:
3119 if fmt.get('targetDurationSec'):
3120 continue
3121
3122 itag = str_or_none(fmt.get('itag'))
3123 audio_track = fmt.get('audioTrack') or {}
3124 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3125 if stream_id in stream_ids:
3126 continue
3127
3128 quality = fmt.get('quality')
3129 height = int_or_none(fmt.get('height'))
3130 if quality == 'tiny' or not quality:
3131 quality = fmt.get('audioQuality', '').lower() or quality
3132 # The 3gp format (17) in android client has a quality of "small",
3133 # but is actually worse than other formats
3134 if itag == '17':
3135 quality = 'tiny'
3136 if quality:
3137 if itag:
3138 itag_qualities[itag] = quality
3139 if height:
3140 res_qualities[height] = quality
3141 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3142 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3143 # number of fragment that would subsequently requested with (`&sq=N`)
3144 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3145 continue
3146
3147 fmt_url = fmt.get('url')
3148 if not fmt_url:
3149 sc = compat_parse_qs(fmt.get('signatureCipher'))
3150 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3151 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3152 if not (sc and fmt_url and encrypted_sig):
3153 continue
3154 if not player_url:
3155 continue
3156 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
3157 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
3158 fmt_url += '&' + sp + '=' + signature
3159
3160 query = parse_qs(fmt_url)
3161 throttled = False
3162 if query.get('n'):
3163 try:
3164 fmt_url = update_url_query(fmt_url, {
3165 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3166 except ExtractorError as e:
3167 self.report_warning(
3168 'nsig extraction failed: You may experience throttling for some formats\n'
3169 f'n = {query["n"][0]} ; player = {player_url}\n{e}', only_once=True)
3170 throttled = True
3171
3172 if itag:
3173 itags[itag] = 'https'
3174 stream_ids.append(stream_id)
3175
3176 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3177 language_preference = (
3178 10 if audio_track.get('audioIsDefault') and 10
3179 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3180 else -1)
3181 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3182 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3183 # Make sure to avoid false positives with small duration differences.
3184 # Eg: __2ABJjxzNo, ySuUZEjARPY
3185 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3186 if is_damaged:
3187 self.report_warning(
3188 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3189 dct = {
3190 'asr': int_or_none(fmt.get('audioSampleRate')),
3191 'filesize': int_or_none(fmt.get('contentLength')),
3192 'format_id': itag,
3193 'format_note': join_nonempty(
3194 '%s%s' % (audio_track.get('displayName') or '',
3195 ' (default)' if language_preference > 0 else ''),
3196 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3197 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3198 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3199 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3200 'fps': int_or_none(fmt.get('fps')) or None,
3201 'height': height,
3202 'quality': q(quality),
3203 'has_drm': bool(fmt.get('drmFamilies')),
3204 'tbr': tbr,
3205 'url': fmt_url,
3206 'width': int_or_none(fmt.get('width')),
3207 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3208 'desc' if language_preference < -1 else ''),
3209 'language_preference': language_preference,
3210 # Strictly de-prioritize damaged and 3gp formats
3211 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3212 }
3213 mime_mobj = re.match(
3214 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3215 if mime_mobj:
3216 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3217 dct.update(parse_codecs(mime_mobj.group(2)))
3218 no_audio = dct.get('acodec') == 'none'
3219 no_video = dct.get('vcodec') == 'none'
3220 if no_audio:
3221 dct['vbr'] = tbr
3222 if no_video:
3223 dct['abr'] = tbr
3224 if no_audio or no_video:
3225 dct['downloader_options'] = {
3226 # Youtube throttles chunks >~10M
3227 'http_chunk_size': 10485760,
3228 }
3229 if dct.get('ext'):
3230 dct['container'] = dct['ext'] + '_dash'
3231 yield dct
3232
3233 live_from_start = is_live and self.get_param('live_from_start')
3234 skip_manifests = self._configuration_arg('skip')
3235 if not self.get_param('youtube_include_hls_manifest', True):
3236 skip_manifests.append('hls')
3237 if not self.get_param('youtube_include_dash_manifest', True):
3238 skip_manifests.append('dash')
3239 get_dash = 'dash' not in skip_manifests and (
3240 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3241 get_hls = not live_from_start and 'hls' not in skip_manifests
3242
3243 def process_manifest_format(f, proto, itag):
3244 if itag in itags:
3245 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3246 return False
3247 itag = f'{itag}-{proto}'
3248 if itag:
3249 f['format_id'] = itag
3250 itags[itag] = proto
3251
3252 f['quality'] = next((
3253 q(qdict[val])
3254 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3255 if val in qdict), -1)
3256 return True
3257
3258 for sd in streaming_data:
3259 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3260 if hls_manifest_url:
3261 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3262 if process_manifest_format(f, 'hls', self._search_regex(
3263 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3264 yield f
3265
3266 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3267 if dash_manifest_url:
3268 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3269 if process_manifest_format(f, 'dash', f['format_id']):
3270 f['filesize'] = int_or_none(self._search_regex(
3271 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3272 if live_from_start:
3273 f['is_from_start'] = True
3274
3275 yield f
3276
3277 def _extract_storyboard(self, player_responses, duration):
3278 spec = get_first(
3279 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3280 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3281 if not base_url:
3282 return
3283 L = len(spec) - 1
3284 for i, args in enumerate(spec):
3285 args = args.split('#')
3286 counts = list(map(int_or_none, args[:5]))
3287 if len(args) != 8 or not all(counts):
3288 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3289 continue
3290 width, height, frame_count, cols, rows = counts
3291 N, sigh = args[6:]
3292
3293 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3294 fragment_count = frame_count / (cols * rows)
3295 fragment_duration = duration / fragment_count
3296 yield {
3297 'format_id': f'sb{i}',
3298 'format_note': 'storyboard',
3299 'ext': 'mhtml',
3300 'protocol': 'mhtml',
3301 'acodec': 'none',
3302 'vcodec': 'none',
3303 'url': url,
3304 'width': width,
3305 'height': height,
3306 'fragments': [{
3307 'url': url.replace('$M', str(j)),
3308 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3309 } for j in range(math.ceil(fragment_count))],
3310 }
3311
3312 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3313 webpage = None
3314 if 'webpage' not in self._configuration_arg('player_skip'):
3315 webpage = self._download_webpage(
3316 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3317
3318 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3319
3320 player_responses, player_url = self._extract_player_responses(
3321 self._get_requested_clients(url, smuggled_data),
3322 video_id, webpage, master_ytcfg)
3323
3324 return webpage, master_ytcfg, player_responses, player_url
3325
3326 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3327 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3328 is_live = get_first(video_details, 'isLive')
3329 if is_live is None:
3330 is_live = get_first(live_broadcast_details, 'isLiveNow')
3331
3332 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3333 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3334
3335 return live_broadcast_details, is_live, streaming_data, formats
3336
3337 def _real_extract(self, url):
3338 url, smuggled_data = unsmuggle_url(url, {})
3339 video_id = self._match_id(url)
3340
3341 base_url = self.http_scheme() + '//www.youtube.com/'
3342 webpage_url = base_url + 'watch?v=' + video_id
3343
3344 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3345
3346 playability_statuses = traverse_obj(
3347 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3348
3349 trailer_video_id = get_first(
3350 playability_statuses,
3351 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3352 expected_type=str)
3353 if trailer_video_id:
3354 return self.url_result(
3355 trailer_video_id, self.ie_key(), trailer_video_id)
3356
3357 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3358 if webpage else (lambda x: None))
3359
3360 video_details = traverse_obj(
3361 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3362 microformats = traverse_obj(
3363 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3364 expected_type=dict, default=[])
3365 video_title = (
3366 get_first(video_details, 'title')
3367 or self._get_text(microformats, (..., 'title'))
3368 or search_meta(['og:title', 'twitter:title', 'title']))
3369 video_description = get_first(video_details, 'shortDescription')
3370
3371 multifeed_metadata_list = get_first(
3372 player_responses,
3373 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3374 expected_type=str)
3375 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3376 if self.get_param('noplaylist'):
3377 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3378 else:
3379 entries = []
3380 feed_ids = []
3381 for feed in multifeed_metadata_list.split(','):
3382 # Unquote should take place before split on comma (,) since textual
3383 # fields may contain comma as well (see
3384 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3385 feed_data = compat_parse_qs(
3386 compat_urllib_parse_unquote_plus(feed))
3387
3388 def feed_entry(name):
3389 return try_get(
3390 feed_data, lambda x: x[name][0], compat_str)
3391
3392 feed_id = feed_entry('id')
3393 if not feed_id:
3394 continue
3395 feed_title = feed_entry('title')
3396 title = video_title
3397 if feed_title:
3398 title += ' (%s)' % feed_title
3399 entries.append({
3400 '_type': 'url_transparent',
3401 'ie_key': 'Youtube',
3402 'url': smuggle_url(
3403 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3404 {'force_singlefeed': True}),
3405 'title': title,
3406 })
3407 feed_ids.append(feed_id)
3408 self.to_screen(
3409 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3410 % (', '.join(feed_ids), video_id))
3411 return self.playlist_result(
3412 entries, video_id, video_title, video_description)
3413
3414 duration = int_or_none(
3415 get_first(video_details, 'lengthSeconds')
3416 or get_first(microformats, 'lengthSeconds')
3417 or parse_duration(search_meta('duration'))) or None
3418
3419 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3420 video_id, microformats, video_details, player_responses, player_url, duration)
3421
3422 if not formats:
3423 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3424 self.report_drm(video_id)
3425 pemr = get_first(
3426 playability_statuses,
3427 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3428 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3429 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3430 if subreason:
3431 if subreason == 'The uploader has not made this video available in your country.':
3432 countries = get_first(microformats, 'availableCountries')
3433 if not countries:
3434 regions_allowed = search_meta('regionsAllowed')
3435 countries = regions_allowed.split(',') if regions_allowed else None
3436 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3437 reason += f'. {subreason}'
3438 if reason:
3439 self.raise_no_formats(reason, expected=True)
3440
3441 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3442 if not keywords and webpage:
3443 keywords = [
3444 unescapeHTML(m.group('content'))
3445 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3446 for keyword in keywords:
3447 if keyword.startswith('yt:stretch='):
3448 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3449 if mobj:
3450 # NB: float is intentional for forcing float division
3451 w, h = (float(v) for v in mobj.groups())
3452 if w > 0 and h > 0:
3453 ratio = w / h
3454 for f in formats:
3455 if f.get('vcodec') != 'none':
3456 f['stretched_ratio'] = ratio
3457 break
3458 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3459 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3460 if thumbnail_url:
3461 thumbnails.append({
3462 'url': thumbnail_url,
3463 })
3464 original_thumbnails = thumbnails.copy()
3465
3466 # The best resolution thumbnails sometimes does not appear in the webpage
3467 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3468 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3469 thumbnail_names = [
3470 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3471 # in resolution, these are not the custom thumbnail. So de-prioritize them
3472 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3473 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3474 ]
3475 n_thumbnail_names = len(thumbnail_names)
3476 thumbnails.extend({
3477 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3478 video_id=video_id, name=name, ext=ext,
3479 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3480 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3481 for thumb in thumbnails:
3482 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3483 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3484 self._remove_duplicate_formats(thumbnails)
3485 self._downloader._sort_thumbnails(original_thumbnails)
3486
3487 category = get_first(microformats, 'category') or search_meta('genre')
3488 channel_id = str_or_none(
3489 get_first(video_details, 'channelId')
3490 or get_first(microformats, 'externalChannelId')
3491 or search_meta('channelId'))
3492 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3493
3494 live_content = get_first(video_details, 'isLiveContent')
3495 is_upcoming = get_first(video_details, 'isUpcoming')
3496 if is_live is None:
3497 if is_upcoming or live_content is False:
3498 is_live = False
3499 if is_upcoming is None and (live_content or is_live):
3500 is_upcoming = False
3501 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3502 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3503 if not duration and live_end_time and live_start_time:
3504 duration = live_end_time - live_start_time
3505
3506 if is_live and self.get_param('live_from_start'):
3507 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3508
3509 formats.extend(self._extract_storyboard(player_responses, duration))
3510
3511 # Source is given priority since formats that throttle are given lower source_preference
3512 # When throttling issue is fully fixed, remove this
3513 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3514
3515 info = {
3516 'id': video_id,
3517 'title': video_title,
3518 'formats': formats,
3519 'thumbnails': thumbnails,
3520 # The best thumbnail that we are sure exists. Prevents unnecessary
3521 # URL checking if user don't care about getting the best possible thumbnail
3522 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3523 'description': video_description,
3524 'uploader': get_first(video_details, 'author'),
3525 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3526 'uploader_url': owner_profile_url,
3527 'channel_id': channel_id,
3528 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3529 'duration': duration,
3530 'view_count': int_or_none(
3531 get_first((video_details, microformats), (..., 'viewCount'))
3532 or search_meta('interactionCount')),
3533 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3534 'age_limit': 18 if (
3535 get_first(microformats, 'isFamilySafe') is False
3536 or search_meta('isFamilyFriendly') == 'false'
3537 or search_meta('og:restrictions:age') == '18+') else 0,
3538 'webpage_url': webpage_url,
3539 'categories': [category] if category else None,
3540 'tags': keywords,
3541 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3542 'is_live': is_live,
3543 'was_live': (False if is_live or is_upcoming or live_content is False
3544 else None if is_live is None or is_upcoming is None
3545 else live_content),
3546 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3547 'release_timestamp': live_start_time,
3548 }
3549
3550 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3551 if pctr:
3552 def get_lang_code(track):
3553 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3554 or track.get('languageCode'))
3555
3556 # Converted into dicts to remove duplicates
3557 captions = {
3558 get_lang_code(sub): sub
3559 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3560 translation_languages = {
3561 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3562 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3563
3564 def process_language(container, base_url, lang_code, sub_name, query):
3565 lang_subs = container.setdefault(lang_code, [])
3566 for fmt in self._SUBTITLE_FORMATS:
3567 query.update({
3568 'fmt': fmt,
3569 })
3570 lang_subs.append({
3571 'ext': fmt,
3572 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3573 'name': sub_name,
3574 })
3575
3576 subtitles, automatic_captions = {}, {}
3577 for lang_code, caption_track in captions.items():
3578 base_url = caption_track.get('baseUrl')
3579 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3580 if not base_url:
3581 continue
3582 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3583 if caption_track.get('kind') != 'asr':
3584 if not lang_code:
3585 continue
3586 process_language(
3587 subtitles, base_url, lang_code, lang_name, {})
3588 if not caption_track.get('isTranslatable'):
3589 continue
3590 for trans_code, trans_name in translation_languages.items():
3591 if not trans_code:
3592 continue
3593 orig_trans_code = trans_code
3594 if caption_track.get('kind') != 'asr':
3595 if 'translated_subs' in self._configuration_arg('skip'):
3596 continue
3597 trans_code += f'-{lang_code}'
3598 trans_name += format_field(lang_name, template=' from %s')
3599 # Add an "-orig" label to the original language so that it can be distinguished.
3600 # The subs are returned without "-orig" as well for compatibility
3601 if lang_code == f'a-{orig_trans_code}':
3602 process_language(
3603 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3604 # Setting tlang=lang returns damaged subtitles.
3605 process_language(automatic_captions, base_url, trans_code, trans_name,
3606 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3607 info['automatic_captions'] = automatic_captions
3608 info['subtitles'] = subtitles
3609
3610 parsed_url = compat_urllib_parse_urlparse(url)
3611 for component in [parsed_url.fragment, parsed_url.query]:
3612 query = compat_parse_qs(component)
3613 for k, v in query.items():
3614 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3615 d_k += '_time'
3616 if d_k not in info and k in s_ks:
3617 info[d_k] = parse_duration(query[k][0])
3618
3619 # Youtube Music Auto-generated description
3620 if video_description:
3621 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)
3622 if mobj:
3623 release_year = mobj.group('release_year')
3624 release_date = mobj.group('release_date')
3625 if release_date:
3626 release_date = release_date.replace('-', '')
3627 if not release_year:
3628 release_year = release_date[:4]
3629 info.update({
3630 'album': mobj.group('album'.strip()),
3631 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3632 'track': mobj.group('track').strip(),
3633 'release_date': release_date,
3634 'release_year': int_or_none(release_year),
3635 })
3636
3637 initial_data = None
3638 if webpage:
3639 initial_data = self._extract_yt_initial_variable(
3640 webpage, self._YT_INITIAL_DATA_RE, video_id,
3641 'yt initial data')
3642 if not initial_data:
3643 query = {'videoId': video_id}
3644 query.update(self._get_checkok_params())
3645 initial_data = self._extract_response(
3646 item_id=video_id, ep='next', fatal=False,
3647 ytcfg=master_ytcfg, query=query,
3648 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3649 note='Downloading initial data API JSON')
3650
3651 try: # This will error if there is no livechat
3652 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3653 except (KeyError, IndexError, TypeError):
3654 pass
3655 else:
3656 info.setdefault('subtitles', {})['live_chat'] = [{
3657 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3658 'video_id': video_id,
3659 'ext': 'json',
3660 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3661 }]
3662
3663 if initial_data:
3664 info['chapters'] = (
3665 self._extract_chapters_from_json(initial_data, duration)
3666 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3667 or None)
3668
3669 contents = traverse_obj(
3670 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3671 expected_type=list, default=[])
3672
3673 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3674 if vpir:
3675 stl = vpir.get('superTitleLink')
3676 if stl:
3677 stl = self._get_text(stl)
3678 if try_get(
3679 vpir,
3680 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3681 info['location'] = stl
3682 else:
3683 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3684 if mobj:
3685 info.update({
3686 'series': mobj.group(1),
3687 'season_number': int(mobj.group(2)),
3688 'episode_number': int(mobj.group(3)),
3689 })
3690 for tlb in (try_get(
3691 vpir,
3692 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3693 list) or []):
3694 tbr = tlb.get('toggleButtonRenderer') or {}
3695 for getter, regex in [(
3696 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3697 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3698 lambda x: x['accessibility'],
3699 lambda x: x['accessibilityData']['accessibilityData'],
3700 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3701 label = (try_get(tbr, getter, dict) or {}).get('label')
3702 if label:
3703 mobj = re.match(regex, label)
3704 if mobj:
3705 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3706 break
3707 sbr_tooltip = try_get(
3708 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3709 if sbr_tooltip:
3710 like_count, dislike_count = sbr_tooltip.split(' / ')
3711 info.update({
3712 'like_count': str_to_int(like_count),
3713 'dislike_count': str_to_int(dislike_count),
3714 })
3715 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3716 if vsir:
3717 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3718 info.update({
3719 'channel': self._get_text(vor, 'title'),
3720 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3721
3722 rows = try_get(
3723 vsir,
3724 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3725 list) or []
3726 multiple_songs = False
3727 for row in rows:
3728 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3729 multiple_songs = True
3730 break
3731 for row in rows:
3732 mrr = row.get('metadataRowRenderer') or {}
3733 mrr_title = mrr.get('title')
3734 if not mrr_title:
3735 continue
3736 mrr_title = self._get_text(mrr, 'title')
3737 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3738 if mrr_title == 'License':
3739 info['license'] = mrr_contents_text
3740 elif not multiple_songs:
3741 if mrr_title == 'Album':
3742 info['album'] = mrr_contents_text
3743 elif mrr_title == 'Artist':
3744 info['artist'] = mrr_contents_text
3745 elif mrr_title == 'Song':
3746 info['track'] = mrr_contents_text
3747
3748 fallbacks = {
3749 'channel': 'uploader',
3750 'channel_id': 'uploader_id',
3751 'channel_url': 'uploader_url',
3752 }
3753
3754 # The upload date for scheduled, live and past live streams / premieres in microformats
3755 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3756 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3757 upload_date = (
3758 unified_strdate(get_first(microformats, 'uploadDate'))
3759 or unified_strdate(search_meta('uploadDate')))
3760 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3761 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3762 info['upload_date'] = upload_date
3763
3764 for to, frm in fallbacks.items():
3765 if not info.get(to):
3766 info[to] = info.get(frm)
3767
3768 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3769 v = info.get(s_k)
3770 if v:
3771 info[d_k] = v
3772
3773 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3774 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3775 is_membersonly = None
3776 is_premium = None
3777 if initial_data and is_private is not None:
3778 is_membersonly = False
3779 is_premium = False
3780 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3781 badge_labels = set()
3782 for content in contents:
3783 if not isinstance(content, dict):
3784 continue
3785 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3786 for badge_label in badge_labels:
3787 if badge_label.lower() == 'members only':
3788 is_membersonly = True
3789 elif badge_label.lower() == 'premium':
3790 is_premium = True
3791 elif badge_label.lower() == 'unlisted':
3792 is_unlisted = True
3793
3794 info['availability'] = self._availability(
3795 is_private=is_private,
3796 needs_premium=is_premium,
3797 needs_subscription=is_membersonly,
3798 needs_auth=info['age_limit'] >= 18,
3799 is_unlisted=None if is_private is None else is_unlisted)
3800
3801 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3802
3803 self.mark_watched(video_id, player_responses)
3804
3805 return info
3806
3807
3808 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3809
3810 @staticmethod
3811 def passthrough_smuggled_data(func):
3812 def _smuggle(entries, smuggled_data):
3813 for entry in entries:
3814 # TODO: Convert URL to music.youtube instead.
3815 # Do we need to passthrough any other smuggled_data?
3816 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3817 yield entry
3818
3819 @functools.wraps(func)
3820 def wrapper(self, url):
3821 url, smuggled_data = unsmuggle_url(url, {})
3822 if self.is_music_url(url):
3823 smuggled_data['is_music_url'] = True
3824 info_dict = func(self, url, smuggled_data)
3825 if smuggled_data and info_dict.get('entries'):
3826 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3827 return info_dict
3828 return wrapper
3829
3830 def _extract_channel_id(self, webpage):
3831 channel_id = self._html_search_meta(
3832 'channelId', webpage, 'channel id', default=None)
3833 if channel_id:
3834 return channel_id
3835 channel_url = self._html_search_meta(
3836 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3837 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3838 'twitter:app:url:googleplay'), webpage, 'channel url')
3839 return self._search_regex(
3840 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3841 channel_url, 'channel id')
3842
3843 @staticmethod
3844 def _extract_basic_item_renderer(item):
3845 # Modified from _extract_grid_item_renderer
3846 known_basic_renderers = (
3847 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3848 )
3849 for key, renderer in item.items():
3850 if not isinstance(renderer, dict):
3851 continue
3852 elif key in known_basic_renderers:
3853 return renderer
3854 elif key.startswith('grid') and key.endswith('Renderer'):
3855 return renderer
3856
3857 def _grid_entries(self, grid_renderer):
3858 for item in grid_renderer['items']:
3859 if not isinstance(item, dict):
3860 continue
3861 renderer = self._extract_basic_item_renderer(item)
3862 if not isinstance(renderer, dict):
3863 continue
3864 title = self._get_text(renderer, 'title')
3865
3866 # playlist
3867 playlist_id = renderer.get('playlistId')
3868 if playlist_id:
3869 yield self.url_result(
3870 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3871 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3872 video_title=title)
3873 continue
3874 # video
3875 video_id = renderer.get('videoId')
3876 if video_id:
3877 yield self._extract_video(renderer)
3878 continue
3879 # channel
3880 channel_id = renderer.get('channelId')
3881 if channel_id:
3882 yield self.url_result(
3883 'https://www.youtube.com/channel/%s' % channel_id,
3884 ie=YoutubeTabIE.ie_key(), video_title=title)
3885 continue
3886 # generic endpoint URL support
3887 ep_url = urljoin('https://www.youtube.com/', try_get(
3888 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3889 compat_str))
3890 if ep_url:
3891 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3892 if ie.suitable(ep_url):
3893 yield self.url_result(
3894 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3895 break
3896
3897 def _music_reponsive_list_entry(self, renderer):
3898 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3899 if video_id:
3900 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3901 ie=YoutubeIE.ie_key(), video_id=video_id)
3902 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3903 if playlist_id:
3904 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3905 if video_id:
3906 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3907 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3908 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3909 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3910 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3911 if browse_id:
3912 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3913 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3914
3915 def _shelf_entries_from_content(self, shelf_renderer):
3916 content = shelf_renderer.get('content')
3917 if not isinstance(content, dict):
3918 return
3919 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3920 if renderer:
3921 # TODO: add support for nested playlists so each shelf is processed
3922 # as separate playlist
3923 # TODO: this includes only first N items
3924 yield from self._grid_entries(renderer)
3925 renderer = content.get('horizontalListRenderer')
3926 if renderer:
3927 # TODO
3928 pass
3929
3930 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3931 ep = try_get(
3932 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3933 compat_str)
3934 shelf_url = urljoin('https://www.youtube.com', ep)
3935 if shelf_url:
3936 # Skipping links to another channels, note that checking for
3937 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3938 # will not work
3939 if skip_channels and '/channels?' in shelf_url:
3940 return
3941 title = self._get_text(shelf_renderer, 'title')
3942 yield self.url_result(shelf_url, video_title=title)
3943 # Shelf may not contain shelf URL, fallback to extraction from content
3944 yield from self._shelf_entries_from_content(shelf_renderer)
3945
3946 def _playlist_entries(self, video_list_renderer):
3947 for content in video_list_renderer['contents']:
3948 if not isinstance(content, dict):
3949 continue
3950 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3951 if not isinstance(renderer, dict):
3952 continue
3953 video_id = renderer.get('videoId')
3954 if not video_id:
3955 continue
3956 yield self._extract_video(renderer)
3957
3958 def _rich_entries(self, rich_grid_renderer):
3959 renderer = try_get(
3960 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3961 video_id = renderer.get('videoId')
3962 if not video_id:
3963 return
3964 yield self._extract_video(renderer)
3965
3966 def _video_entry(self, video_renderer):
3967 video_id = video_renderer.get('videoId')
3968 if video_id:
3969 return self._extract_video(video_renderer)
3970
3971 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3972 url = urljoin('https://youtube.com', traverse_obj(
3973 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3974 if url:
3975 return self.url_result(
3976 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3977
3978 def _post_thread_entries(self, post_thread_renderer):
3979 post_renderer = try_get(
3980 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3981 if not post_renderer:
3982 return
3983 # video attachment
3984 video_renderer = try_get(
3985 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3986 video_id = video_renderer.get('videoId')
3987 if video_id:
3988 entry = self._extract_video(video_renderer)
3989 if entry:
3990 yield entry
3991 # playlist attachment
3992 playlist_id = try_get(
3993 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3994 if playlist_id:
3995 yield self.url_result(
3996 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3997 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3998 # inline video links
3999 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4000 for run in runs:
4001 if not isinstance(run, dict):
4002 continue
4003 ep_url = try_get(
4004 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
4005 if not ep_url:
4006 continue
4007 if not YoutubeIE.suitable(ep_url):
4008 continue
4009 ep_video_id = YoutubeIE._match_id(ep_url)
4010 if video_id == ep_video_id:
4011 continue
4012 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4013
4014 def _post_thread_continuation_entries(self, post_thread_continuation):
4015 contents = post_thread_continuation.get('contents')
4016 if not isinstance(contents, list):
4017 return
4018 for content in contents:
4019 renderer = content.get('backstagePostThreadRenderer')
4020 if not isinstance(renderer, dict):
4021 continue
4022 yield from self._post_thread_entries(renderer)
4023
4024 r''' # unused
4025 def _rich_grid_entries(self, contents):
4026 for content in contents:
4027 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4028 if video_renderer:
4029 entry = self._video_entry(video_renderer)
4030 if entry:
4031 yield entry
4032 '''
4033
4034 def _extract_entries(self, parent_renderer, continuation_list):
4035 # continuation_list is modified in-place with continuation_list = [continuation_token]
4036 continuation_list[:] = [None]
4037 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4038 for content in contents:
4039 if not isinstance(content, dict):
4040 continue
4041 is_renderer = traverse_obj(
4042 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4043 expected_type=dict)
4044 if not is_renderer:
4045 renderer = content.get('richItemRenderer')
4046 if renderer:
4047 for entry in self._rich_entries(renderer):
4048 yield entry
4049 continuation_list[0] = self._extract_continuation(parent_renderer)
4050 continue
4051 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4052 for isr_content in isr_contents:
4053 if not isinstance(isr_content, dict):
4054 continue
4055
4056 known_renderers = {
4057 'playlistVideoListRenderer': self._playlist_entries,
4058 'gridRenderer': self._grid_entries,
4059 'reelShelfRenderer': self._grid_entries,
4060 'shelfRenderer': self._shelf_entries,
4061 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4062 'backstagePostThreadRenderer': self._post_thread_entries,
4063 'videoRenderer': lambda x: [self._video_entry(x)],
4064 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4065 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4066 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4067 }
4068 for key, renderer in isr_content.items():
4069 if key not in known_renderers:
4070 continue
4071 for entry in known_renderers[key](renderer):
4072 if entry:
4073 yield entry
4074 continuation_list[0] = self._extract_continuation(renderer)
4075 break
4076
4077 if not continuation_list[0]:
4078 continuation_list[0] = self._extract_continuation(is_renderer)
4079
4080 if not continuation_list[0]:
4081 continuation_list[0] = self._extract_continuation(parent_renderer)
4082
4083 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4084 continuation_list = [None]
4085 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4086 tab_content = try_get(tab, lambda x: x['content'], dict)
4087 if not tab_content:
4088 return
4089 parent_renderer = (
4090 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4091 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4092 yield from extract_entries(parent_renderer)
4093 continuation = continuation_list[0]
4094
4095 for page_num in itertools.count(1):
4096 if not continuation:
4097 break
4098 headers = self.generate_api_headers(
4099 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4100 response = self._extract_response(
4101 item_id=f'{item_id} page {page_num}',
4102 query=continuation, headers=headers, ytcfg=ytcfg,
4103 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4104
4105 if not response:
4106 break
4107 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4108 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4109 visitor_data = self._extract_visitor_data(response) or visitor_data
4110
4111 known_continuation_renderers = {
4112 'playlistVideoListContinuation': self._playlist_entries,
4113 'gridContinuation': self._grid_entries,
4114 'itemSectionContinuation': self._post_thread_continuation_entries,
4115 'sectionListContinuation': extract_entries, # for feeds
4116 }
4117 continuation_contents = try_get(
4118 response, lambda x: x['continuationContents'], dict) or {}
4119 continuation_renderer = None
4120 for key, value in continuation_contents.items():
4121 if key not in known_continuation_renderers:
4122 continue
4123 continuation_renderer = value
4124 continuation_list = [None]
4125 yield from known_continuation_renderers[key](continuation_renderer)
4126 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4127 break
4128 if continuation_renderer:
4129 continue
4130
4131 known_renderers = {
4132 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4133 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4134 'gridVideoRenderer': (self._grid_entries, 'items'),
4135 'gridChannelRenderer': (self._grid_entries, 'items'),
4136 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4137 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4138 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4139 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4140 }
4141 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4142 continuation_items = try_get(
4143 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4144 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4145 video_items_renderer = None
4146 for key, value in continuation_item.items():
4147 if key not in known_renderers:
4148 continue
4149 video_items_renderer = {known_renderers[key][1]: continuation_items}
4150 continuation_list = [None]
4151 yield from known_renderers[key][0](video_items_renderer)
4152 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4153 break
4154 if video_items_renderer:
4155 continue
4156 break
4157
4158 @staticmethod
4159 def _extract_selected_tab(tabs, fatal=True):
4160 for tab in tabs:
4161 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4162 if renderer.get('selected') is True:
4163 return renderer
4164 else:
4165 if fatal:
4166 raise ExtractorError('Unable to find selected tab')
4167
4168 def _extract_uploader(self, data):
4169 uploader = {}
4170 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4171 owner = try_get(
4172 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4173 if owner:
4174 owner_text = owner.get('text')
4175 uploader['uploader'] = self._search_regex(
4176 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4177 uploader['uploader_id'] = try_get(
4178 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4179 uploader['uploader_url'] = urljoin(
4180 'https://www.youtube.com/',
4181 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4182 return {k: v for k, v in uploader.items() if v is not None}
4183
4184 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4185 playlist_id = title = description = channel_url = channel_name = channel_id = None
4186 tags = []
4187
4188 selected_tab = self._extract_selected_tab(tabs)
4189 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4190 renderer = try_get(
4191 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4192 if renderer:
4193 channel_name = renderer.get('title')
4194 channel_url = renderer.get('channelUrl')
4195 channel_id = renderer.get('externalId')
4196 else:
4197 renderer = try_get(
4198 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4199
4200 if renderer:
4201 title = renderer.get('title')
4202 description = renderer.get('description', '')
4203 playlist_id = channel_id
4204 tags = renderer.get('keywords', '').split()
4205
4206 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4207 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4208 def _get_uncropped(url):
4209 return url_or_none((url or '').split('=')[0] + '=s0')
4210
4211 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4212 if avatar_thumbnails:
4213 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4214 if uncropped_avatar:
4215 avatar_thumbnails.append({
4216 'url': uncropped_avatar,
4217 'id': 'avatar_uncropped',
4218 'preference': 1
4219 })
4220
4221 channel_banners = self._extract_thumbnails(
4222 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4223 for banner in channel_banners:
4224 banner['preference'] = -10
4225
4226 if channel_banners:
4227 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4228 if uncropped_banner:
4229 channel_banners.append({
4230 'url': uncropped_banner,
4231 'id': 'banner_uncropped',
4232 'preference': -5
4233 })
4234
4235 primary_thumbnails = self._extract_thumbnails(
4236 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4237
4238 if playlist_id is None:
4239 playlist_id = item_id
4240
4241 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4242 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4243 if title is None:
4244 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4245 title += format_field(selected_tab, 'title', ' - %s')
4246 title += format_field(selected_tab, 'expandedText', ' - %s')
4247
4248 metadata = {
4249 'playlist_id': playlist_id,
4250 'playlist_title': title,
4251 'playlist_description': description,
4252 'uploader': channel_name,
4253 'uploader_id': channel_id,
4254 'uploader_url': channel_url,
4255 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4256 'tags': tags,
4257 'view_count': self._get_count(playlist_stats, 1),
4258 'availability': self._extract_availability(data),
4259 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4260 'playlist_count': self._get_count(playlist_stats, 0),
4261 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4262 }
4263 if not channel_id:
4264 metadata.update(self._extract_uploader(data))
4265 metadata.update({
4266 'channel': metadata['uploader'],
4267 'channel_id': metadata['uploader_id'],
4268 'channel_url': metadata['uploader_url']})
4269 return self.playlist_result(
4270 self._entries(
4271 selected_tab, playlist_id, ytcfg,
4272 self._extract_account_syncid(ytcfg, data),
4273 self._extract_visitor_data(data, ytcfg)),
4274 **metadata)
4275
4276 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4277 first_id = last_id = response = None
4278 for page_num in itertools.count(1):
4279 videos = list(self._playlist_entries(playlist))
4280 if not videos:
4281 return
4282 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4283 if start >= len(videos):
4284 return
4285 for video in videos[start:]:
4286 yield video
4287 first_id = first_id or videos[0]['id']
4288 last_id = videos[-1]['id']
4289 watch_endpoint = try_get(
4290 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4291 headers = self.generate_api_headers(
4292 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4293 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4294 query = {
4295 'playlistId': playlist_id,
4296 'videoId': watch_endpoint.get('videoId') or last_id,
4297 'index': watch_endpoint.get('index') or len(videos),
4298 'params': watch_endpoint.get('params') or 'OAE%3D'
4299 }
4300 response = self._extract_response(
4301 item_id='%s page %d' % (playlist_id, page_num),
4302 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4303 check_get_keys='contents'
4304 )
4305 playlist = try_get(
4306 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4307
4308 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4309 title = playlist.get('title') or try_get(
4310 data, lambda x: x['titleText']['simpleText'], compat_str)
4311 playlist_id = playlist.get('playlistId') or item_id
4312
4313 # Delegating everything except mix playlists to regular tab-based playlist URL
4314 playlist_url = urljoin(url, try_get(
4315 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4316 compat_str))
4317
4318 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4319 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4320 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4321
4322 if playlist_url and playlist_url != url and not is_known_unviewable:
4323 return self.url_result(
4324 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4325 video_title=title)
4326
4327 return self.playlist_result(
4328 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4329 playlist_id=playlist_id, playlist_title=title)
4330
4331 def _extract_availability(self, data):
4332 """
4333 Gets the availability of a given playlist/tab.
4334 Note: Unless YouTube tells us explicitly, we do not assume it is public
4335 @param data: response
4336 """
4337 is_private = is_unlisted = None
4338 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4339 badge_labels = self._extract_badges(renderer)
4340
4341 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4342 privacy_dropdown_entries = try_get(
4343 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4344 for renderer_dict in privacy_dropdown_entries:
4345 is_selected = try_get(
4346 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4347 if not is_selected:
4348 continue
4349 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4350 if label:
4351 badge_labels.add(label.lower())
4352 break
4353
4354 for badge_label in badge_labels:
4355 if badge_label == 'unlisted':
4356 is_unlisted = True
4357 elif badge_label == 'private':
4358 is_private = True
4359 elif badge_label == 'public':
4360 is_unlisted = is_private = False
4361 return self._availability(is_private, False, False, False, is_unlisted)
4362
4363 @staticmethod
4364 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4365 sidebar_renderer = try_get(
4366 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4367 for item in sidebar_renderer:
4368 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4369 if renderer:
4370 return renderer
4371
4372 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4373 """
4374 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4375 """
4376 browse_id = params = None
4377 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4378 if not renderer:
4379 return
4380 menu_renderer = try_get(
4381 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4382 for menu_item in menu_renderer:
4383 if not isinstance(menu_item, dict):
4384 continue
4385 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4386 text = try_get(
4387 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4388 if not text or text.lower() != 'show unavailable videos':
4389 continue
4390 browse_endpoint = try_get(
4391 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4392 browse_id = browse_endpoint.get('browseId')
4393 params = browse_endpoint.get('params')
4394 break
4395
4396 headers = self.generate_api_headers(
4397 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4398 visitor_data=self._extract_visitor_data(data, ytcfg))
4399 query = {
4400 'params': params or 'wgYCCAA=',
4401 'browseId': browse_id or 'VL%s' % item_id
4402 }
4403 return self._extract_response(
4404 item_id=item_id, headers=headers, query=query,
4405 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4406 note='Downloading API JSON with unavailable videos')
4407
4408 @property
4409 def skip_webpage(self):
4410 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4411
4412 def _extract_webpage(self, url, item_id, fatal=True):
4413 retries = self.get_param('extractor_retries', 3)
4414 count = -1
4415 webpage = data = last_error = None
4416 while count < retries:
4417 count += 1
4418 # Sometimes youtube returns a webpage with incomplete ytInitialData
4419 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4420 if last_error:
4421 self.report_warning('%s. Retrying ...' % last_error)
4422 try:
4423 webpage = self._download_webpage(
4424 url, item_id,
4425 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4426 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4427 except ExtractorError as e:
4428 if isinstance(e.cause, network_exceptions):
4429 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4430 last_error = error_to_compat_str(e.cause or e.msg)
4431 if count < retries:
4432 continue
4433 if fatal:
4434 raise
4435 self.report_warning(error_to_compat_str(e))
4436 break
4437 else:
4438 try:
4439 self._extract_and_report_alerts(data)
4440 except ExtractorError as e:
4441 if fatal:
4442 raise
4443 self.report_warning(error_to_compat_str(e))
4444 break
4445
4446 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4447 break
4448
4449 last_error = 'Incomplete yt initial data received'
4450 if count >= retries:
4451 if fatal:
4452 raise ExtractorError(last_error)
4453 self.report_warning(last_error)
4454 break
4455
4456 return webpage, data
4457
4458 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4459 """Use if failed to extract ytcfg (and data) from initial webpage"""
4460 if not ytcfg and self.is_authenticated:
4461 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4462 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4463 raise ExtractorError(
4464 f'{msg}. If you are not downloading private content, or '
4465 'your cookies are only for the first account and channel,'
4466 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4467 expected=True)
4468 self.report_warning(msg, only_once=True)
4469
4470 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4471 data = None
4472 if not self.skip_webpage:
4473 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4474 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4475 # Reject webpage data if redirected to home page without explicitly requesting
4476 selected_tab = self._extract_selected_tab(traverse_obj(
4477 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4478 if (url != 'https://www.youtube.com/feed/recommended'
4479 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4480 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4481 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4482 if fatal:
4483 raise ExtractorError(msg, expected=True)
4484 self.report_warning(msg, only_once=True)
4485 if not data:
4486 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4487 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4488 return data, ytcfg
4489
4490 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4491 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4492 resolve_response = self._extract_response(
4493 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4494 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4495 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4496 for ep_key, ep in endpoints.items():
4497 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4498 if params:
4499 return self._extract_response(
4500 item_id=item_id, query=params, ep=ep, headers=headers,
4501 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4502 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4503 err_note = 'Failed to resolve url (does the playlist exist?)'
4504 if fatal:
4505 raise ExtractorError(err_note, expected=True)
4506 self.report_warning(err_note, item_id)
4507
4508 _SEARCH_PARAMS = None
4509
4510 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4511 data = {'query': query}
4512 if params is NO_DEFAULT:
4513 params = self._SEARCH_PARAMS
4514 if params:
4515 data['params'] = params
4516
4517 content_keys = (
4518 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4519 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4520 # ytmusic search
4521 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4522 ('continuationContents', ),
4523 )
4524 display_id = f'query "{query}"'
4525 check_get_keys = tuple({keys[0] for keys in content_keys})
4526 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4527 self._report_playlist_authcheck(ytcfg, fatal=False)
4528
4529 continuation_list = [None]
4530 search = None
4531 for page_num in itertools.count(1):
4532 data.update(continuation_list[0] or {})
4533 headers = self.generate_api_headers(
4534 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4535 search = self._extract_response(
4536 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4537 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4538 slr_contents = traverse_obj(search, *content_keys)
4539 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4540 if not continuation_list[0]:
4541 break
4542
4543
4544 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4545 IE_DESC = 'YouTube Tabs'
4546 _VALID_URL = r'''(?x:
4547 https?://
4548 (?:\w+\.)?
4549 (?:
4550 youtube(?:kids)?\.com|
4551 %(invidious)s
4552 )/
4553 (?:
4554 (?P<channel_type>channel|c|user|browse)/|
4555 (?P<not_channel>
4556 feed/|hashtag/|
4557 (?:playlist|watch)\?.*?\blist=
4558 )|
4559 (?!(?:%(reserved_names)s)\b) # Direct URLs
4560 )
4561 (?P<id>[^/?\#&]+)
4562 )''' % {
4563 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4564 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4565 }
4566 IE_NAME = 'youtube:tab'
4567
4568 _TESTS = [{
4569 'note': 'playlists, multipage',
4570 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4571 'playlist_mincount': 94,
4572 'info_dict': {
4573 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4574 'title': 'Igor Kleiner - Playlists',
4575 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4576 'uploader': 'Igor Kleiner',
4577 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4578 'channel': 'Igor Kleiner',
4579 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4580 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4581 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4582 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4583 'channel_follower_count': int
4584 },
4585 }, {
4586 'note': 'playlists, multipage, different order',
4587 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4588 'playlist_mincount': 94,
4589 'info_dict': {
4590 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4591 'title': 'Igor Kleiner - Playlists',
4592 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4593 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4594 'uploader': 'Igor Kleiner',
4595 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4596 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4597 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4598 'channel': 'Igor Kleiner',
4599 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4600 'channel_follower_count': int
4601 },
4602 }, {
4603 'note': 'playlists, series',
4604 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4605 'playlist_mincount': 5,
4606 'info_dict': {
4607 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4608 'title': '3Blue1Brown - Playlists',
4609 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4610 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4611 'uploader': '3Blue1Brown',
4612 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4613 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4614 'channel': '3Blue1Brown',
4615 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4616 'tags': ['Mathematics'],
4617 'channel_follower_count': int
4618 },
4619 }, {
4620 'note': 'playlists, singlepage',
4621 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4622 'playlist_mincount': 4,
4623 'info_dict': {
4624 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4625 'title': 'ThirstForScience - Playlists',
4626 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4627 'uploader': 'ThirstForScience',
4628 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4629 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4630 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4631 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4632 'tags': 'count:13',
4633 'channel': 'ThirstForScience',
4634 'channel_follower_count': int
4635 }
4636 }, {
4637 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4638 'only_matching': True,
4639 }, {
4640 'note': 'basic, single video playlist',
4641 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4642 'info_dict': {
4643 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4644 'uploader': 'Sergey M.',
4645 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4646 'title': 'youtube-dl public playlist',
4647 'description': '',
4648 'tags': [],
4649 'view_count': int,
4650 'modified_date': '20201130',
4651 'channel': 'Sergey M.',
4652 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4653 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4654 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4655 },
4656 'playlist_count': 1,
4657 }, {
4658 'note': 'empty playlist',
4659 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4660 'info_dict': {
4661 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4662 'uploader': 'Sergey M.',
4663 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4664 'title': 'youtube-dl empty playlist',
4665 'tags': [],
4666 'channel': 'Sergey M.',
4667 'description': '',
4668 'modified_date': '20160902',
4669 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4670 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4671 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4672 },
4673 'playlist_count': 0,
4674 }, {
4675 'note': 'Home tab',
4676 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4677 'info_dict': {
4678 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4679 'title': 'lex will - Home',
4680 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4681 'uploader': 'lex will',
4682 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4683 'channel': 'lex will',
4684 'tags': ['bible', 'history', 'prophesy'],
4685 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4686 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4687 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4688 'channel_follower_count': int
4689 },
4690 'playlist_mincount': 2,
4691 }, {
4692 'note': 'Videos tab',
4693 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4694 'info_dict': {
4695 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4696 'title': 'lex will - Videos',
4697 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4698 'uploader': 'lex will',
4699 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4700 'tags': ['bible', 'history', 'prophesy'],
4701 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4702 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4703 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4704 'channel': 'lex will',
4705 'channel_follower_count': int
4706 },
4707 'playlist_mincount': 975,
4708 }, {
4709 'note': 'Videos tab, sorted by popular',
4710 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4711 'info_dict': {
4712 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4713 'title': 'lex will - Videos',
4714 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4715 'uploader': 'lex will',
4716 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4717 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4718 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4719 'channel': 'lex will',
4720 'tags': ['bible', 'history', 'prophesy'],
4721 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4722 'channel_follower_count': int
4723 },
4724 'playlist_mincount': 199,
4725 }, {
4726 'note': 'Playlists tab',
4727 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4728 'info_dict': {
4729 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4730 'title': 'lex will - Playlists',
4731 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4732 'uploader': 'lex will',
4733 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4734 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4735 'channel': 'lex will',
4736 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4737 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4738 'tags': ['bible', 'history', 'prophesy'],
4739 'channel_follower_count': int
4740 },
4741 'playlist_mincount': 17,
4742 }, {
4743 'note': 'Community tab',
4744 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4745 'info_dict': {
4746 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4747 'title': 'lex will - Community',
4748 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4749 'uploader': 'lex will',
4750 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4751 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4752 'channel': 'lex will',
4753 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4754 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4755 'tags': ['bible', 'history', 'prophesy'],
4756 'channel_follower_count': int
4757 },
4758 'playlist_mincount': 18,
4759 }, {
4760 'note': 'Channels tab',
4761 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4762 'info_dict': {
4763 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4764 'title': 'lex will - Channels',
4765 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4766 'uploader': 'lex will',
4767 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4768 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4769 'channel': 'lex will',
4770 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4771 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4772 'tags': ['bible', 'history', 'prophesy'],
4773 'channel_follower_count': int
4774 },
4775 'playlist_mincount': 12,
4776 }, {
4777 'note': 'Search tab',
4778 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4779 'playlist_mincount': 40,
4780 'info_dict': {
4781 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4782 'title': '3Blue1Brown - Search - linear algebra',
4783 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4784 'uploader': '3Blue1Brown',
4785 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4786 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4787 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4788 'tags': ['Mathematics'],
4789 'channel': '3Blue1Brown',
4790 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4791 'channel_follower_count': int
4792 },
4793 }, {
4794 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4795 'only_matching': True,
4796 }, {
4797 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4798 'only_matching': True,
4799 }, {
4800 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4801 'only_matching': True,
4802 }, {
4803 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4804 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4805 'info_dict': {
4806 'title': '29C3: Not my department',
4807 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4808 'uploader': 'Christiaan008',
4809 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4810 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4811 'tags': [],
4812 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4813 'view_count': int,
4814 'modified_date': '20150605',
4815 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4816 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4817 'channel': 'Christiaan008',
4818 },
4819 'playlist_count': 96,
4820 }, {
4821 'note': 'Large playlist',
4822 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4823 'info_dict': {
4824 'title': 'Uploads from Cauchemar',
4825 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4826 'uploader': 'Cauchemar',
4827 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4828 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4829 'tags': [],
4830 'modified_date': r're:\d{8}',
4831 'channel': 'Cauchemar',
4832 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4833 'view_count': int,
4834 'description': '',
4835 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4836 },
4837 'playlist_mincount': 1123,
4838 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4839 }, {
4840 'note': 'even larger playlist, 8832 videos',
4841 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4842 'only_matching': True,
4843 }, {
4844 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4845 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4846 'info_dict': {
4847 'title': 'Uploads from Interstellar Movie',
4848 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4849 'uploader': 'Interstellar Movie',
4850 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4851 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4852 'tags': [],
4853 'view_count': int,
4854 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4855 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4856 'channel': 'Interstellar Movie',
4857 'description': '',
4858 'modified_date': r're:\d{8}',
4859 },
4860 'playlist_mincount': 21,
4861 }, {
4862 'note': 'Playlist with "show unavailable videos" button',
4863 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4864 'info_dict': {
4865 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4866 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4867 'uploader': 'Phim Siêu Nhân Nhật Bản',
4868 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4869 'view_count': int,
4870 'channel': 'Phim Siêu Nhân Nhật Bản',
4871 'tags': [],
4872 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4873 'description': '',
4874 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4875 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4876 'modified_date': r're:\d{8}',
4877 },
4878 'playlist_mincount': 200,
4879 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4880 }, {
4881 'note': 'Playlist with unavailable videos in page 7',
4882 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4883 'info_dict': {
4884 'title': 'Uploads from BlankTV',
4885 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4886 'uploader': 'BlankTV',
4887 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4888 'channel': 'BlankTV',
4889 'channel_url': 'https://www.youtube.com/c/blanktv',
4890 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4891 'view_count': int,
4892 'tags': [],
4893 'uploader_url': 'https://www.youtube.com/c/blanktv',
4894 'modified_date': r're:\d{8}',
4895 'description': '',
4896 },
4897 'playlist_mincount': 1000,
4898 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4899 }, {
4900 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4901 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4902 'info_dict': {
4903 'title': 'Data Analysis with Dr Mike Pound',
4904 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4905 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4906 'uploader': 'Computerphile',
4907 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4908 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4909 'tags': [],
4910 'view_count': int,
4911 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4912 'channel_url': 'https://www.youtube.com/user/Computerphile',
4913 'channel': 'Computerphile',
4914 },
4915 'playlist_mincount': 11,
4916 }, {
4917 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4918 'only_matching': True,
4919 }, {
4920 'note': 'Playlist URL that does not actually serve a playlist',
4921 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4922 'info_dict': {
4923 'id': 'FqZTN594JQw',
4924 'ext': 'webm',
4925 'title': "Smiley's People 01 detective, Adventure Series, Action",
4926 'uploader': 'STREEM',
4927 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4928 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4929 'upload_date': '20150526',
4930 'license': 'Standard YouTube License',
4931 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4932 'categories': ['People & Blogs'],
4933 'tags': list,
4934 'view_count': int,
4935 'like_count': int,
4936 },
4937 'params': {
4938 'skip_download': True,
4939 },
4940 'skip': 'This video is not available.',
4941 'add_ie': [YoutubeIE.ie_key()],
4942 }, {
4943 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4944 'only_matching': True,
4945 }, {
4946 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4947 'only_matching': True,
4948 }, {
4949 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4950 'info_dict': {
4951 'id': 'GgL890LIznQ', # This will keep changing
4952 'ext': 'mp4',
4953 'title': str,
4954 'uploader': 'Sky News',
4955 'uploader_id': 'skynews',
4956 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4957 'upload_date': r're:\d{8}',
4958 'description': str,
4959 'categories': ['News & Politics'],
4960 'tags': list,
4961 'like_count': int,
4962 'release_timestamp': 1642502819,
4963 'channel': 'Sky News',
4964 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4965 'age_limit': 0,
4966 'view_count': int,
4967 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4968 'playable_in_embed': True,
4969 'release_date': '20220118',
4970 'availability': 'public',
4971 'live_status': 'is_live',
4972 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4973 'channel_follower_count': int
4974 },
4975 'params': {
4976 'skip_download': True,
4977 },
4978 'expected_warnings': ['Ignoring subtitle tracks found in '],
4979 }, {
4980 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4981 'info_dict': {
4982 'id': 'a48o2S1cPoo',
4983 'ext': 'mp4',
4984 'title': 'The Young Turks - Live Main Show',
4985 'uploader': 'The Young Turks',
4986 'uploader_id': 'TheYoungTurks',
4987 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4988 'upload_date': '20150715',
4989 'license': 'Standard YouTube License',
4990 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4991 'categories': ['News & Politics'],
4992 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4993 'like_count': int,
4994 },
4995 'params': {
4996 'skip_download': True,
4997 },
4998 'only_matching': True,
4999 }, {
5000 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5001 'only_matching': True,
5002 }, {
5003 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5004 'only_matching': True,
5005 }, {
5006 'note': 'A channel that is not live. Should raise error',
5007 'url': 'https://www.youtube.com/user/numberphile/live',
5008 'only_matching': True,
5009 }, {
5010 'url': 'https://www.youtube.com/feed/trending',
5011 'only_matching': True,
5012 }, {
5013 'url': 'https://www.youtube.com/feed/library',
5014 'only_matching': True,
5015 }, {
5016 'url': 'https://www.youtube.com/feed/history',
5017 'only_matching': True,
5018 }, {
5019 'url': 'https://www.youtube.com/feed/subscriptions',
5020 'only_matching': True,
5021 }, {
5022 'url': 'https://www.youtube.com/feed/watch_later',
5023 'only_matching': True,
5024 }, {
5025 'note': 'Recommended - redirects to home page.',
5026 'url': 'https://www.youtube.com/feed/recommended',
5027 'only_matching': True,
5028 }, {
5029 'note': 'inline playlist with not always working continuations',
5030 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5031 'only_matching': True,
5032 }, {
5033 'url': 'https://www.youtube.com/course',
5034 'only_matching': True,
5035 }, {
5036 'url': 'https://www.youtube.com/zsecurity',
5037 'only_matching': True,
5038 }, {
5039 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5040 'only_matching': True,
5041 }, {
5042 'url': 'https://www.youtube.com/TheYoungTurks/live',
5043 'only_matching': True,
5044 }, {
5045 'url': 'https://www.youtube.com/hashtag/cctv9',
5046 'info_dict': {
5047 'id': 'cctv9',
5048 'title': '#cctv9',
5049 'tags': [],
5050 },
5051 'playlist_mincount': 350,
5052 }, {
5053 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5054 'only_matching': True,
5055 }, {
5056 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5057 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5058 'only_matching': True
5059 }, {
5060 'note': '/browse/ should redirect to /channel/',
5061 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5062 'only_matching': True
5063 }, {
5064 'note': 'VLPL, should redirect to playlist?list=PL...',
5065 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5066 'info_dict': {
5067 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5068 'uploader': 'NoCopyrightSounds',
5069 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5070 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5071 'title': 'NCS Releases',
5072 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5073 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5074 'modified_date': r're:\d{8}',
5075 'view_count': int,
5076 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5077 'tags': [],
5078 'channel': 'NoCopyrightSounds',
5079 },
5080 'playlist_mincount': 166,
5081 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5082 }, {
5083 'note': 'Topic, should redirect to playlist?list=UU...',
5084 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5085 'info_dict': {
5086 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5087 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5088 'title': 'Uploads from Royalty Free Music - Topic',
5089 'uploader': 'Royalty Free Music - Topic',
5090 'tags': [],
5091 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5092 'channel': 'Royalty Free Music - Topic',
5093 'view_count': int,
5094 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5095 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5096 'modified_date': r're:\d{8}',
5097 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5098 'description': '',
5099 },
5100 'expected_warnings': [
5101 'The URL does not have a videos tab',
5102 r'[Uu]navailable videos (are|will be) hidden',
5103 ],
5104 'playlist_mincount': 101,
5105 }, {
5106 'note': 'Topic without a UU playlist',
5107 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5108 'info_dict': {
5109 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5110 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5111 'tags': [],
5112 },
5113 'expected_warnings': [
5114 'the playlist redirect gave error',
5115 ],
5116 'playlist_mincount': 9,
5117 }, {
5118 'note': 'Youtube music Album',
5119 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5120 'info_dict': {
5121 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5122 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5123 'tags': [],
5124 'view_count': int,
5125 'description': '',
5126 'availability': 'unlisted',
5127 'modified_date': r're:\d{8}',
5128 },
5129 'playlist_count': 50,
5130 }, {
5131 'note': 'unlisted single video playlist',
5132 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5133 'info_dict': {
5134 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5135 'uploader': 'colethedj',
5136 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5137 'title': 'yt-dlp unlisted playlist test',
5138 'availability': 'unlisted',
5139 'tags': [],
5140 'modified_date': '20211208',
5141 'channel': 'colethedj',
5142 'view_count': int,
5143 'description': '',
5144 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5145 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5146 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5147 },
5148 'playlist_count': 1,
5149 }, {
5150 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5151 'url': 'https://www.youtube.com/feed/recommended',
5152 'info_dict': {
5153 'id': 'recommended',
5154 'title': 'recommended',
5155 'tags': [],
5156 },
5157 'playlist_mincount': 50,
5158 'params': {
5159 'skip_download': True,
5160 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5161 },
5162 }, {
5163 'note': 'API Fallback: /videos tab, sorted by oldest first',
5164 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5165 'info_dict': {
5166 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5167 'title': 'Cody\'sLab - Videos',
5168 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5169 'uploader': 'Cody\'sLab',
5170 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5171 'channel': 'Cody\'sLab',
5172 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5173 'tags': [],
5174 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5175 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5176 'channel_follower_count': int
5177 },
5178 'playlist_mincount': 650,
5179 'params': {
5180 'skip_download': True,
5181 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5182 },
5183 }, {
5184 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5185 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5186 'info_dict': {
5187 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5188 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5189 'title': 'Uploads from Royalty Free Music - Topic',
5190 'uploader': 'Royalty Free Music - Topic',
5191 'modified_date': r're:\d{8}',
5192 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5193 'description': '',
5194 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5195 'tags': [],
5196 'channel': 'Royalty Free Music - Topic',
5197 'view_count': int,
5198 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5199 },
5200 'expected_warnings': [
5201 'does not have a videos tab',
5202 r'[Uu]navailable videos (are|will be) hidden',
5203 ],
5204 'playlist_mincount': 101,
5205 'params': {
5206 'skip_download': True,
5207 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5208 },
5209 }, {
5210 'note': 'non-standard redirect to regional channel',
5211 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5212 'only_matching': True
5213 }, {
5214 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5215 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5216 'info_dict': {
5217 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5218 'modified_date': '20220407',
5219 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5220 'tags': [],
5221 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5222 'uploader': 'pukkandan',
5223 'availability': 'unlisted',
5224 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5225 'channel': 'pukkandan',
5226 'description': 'Test for collaborative playlist',
5227 'title': 'yt-dlp test - collaborative playlist',
5228 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5229 },
5230 'playlist_mincount': 2
5231 }]
5232
5233 @classmethod
5234 def suitable(cls, url):
5235 return False if YoutubeIE.suitable(url) else super().suitable(url)
5236
5237 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5238
5239 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5240 def _real_extract(self, url, smuggled_data):
5241 item_id = self._match_id(url)
5242 url = compat_urlparse.urlunparse(
5243 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5244 compat_opts = self.get_param('compat_opts', [])
5245
5246 def get_mobj(url):
5247 mobj = self._URL_RE.match(url).groupdict()
5248 mobj.update((k, '') for k, v in mobj.items() if v is None)
5249 return mobj
5250
5251 mobj, redirect_warning = get_mobj(url), None
5252 # Youtube returns incomplete data if tabname is not lower case
5253 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5254 if is_channel:
5255 if smuggled_data.get('is_music_url'):
5256 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5257 item_id = item_id[2:]
5258 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5259 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5260 mdata = self._extract_tab_endpoint(
5261 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5262 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5263 get_all=False, expected_type=compat_str)
5264 if not murl:
5265 raise ExtractorError('Failed to resolve album to playlist')
5266 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5267 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5268 pre = f'https://www.youtube.com/channel/{item_id}'
5269
5270 original_tab_name = tab
5271 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5272 # Home URLs should redirect to /videos/
5273 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5274 'To download only the videos in the home page, add a "/featured" to the URL')
5275 tab = '/videos'
5276
5277 url = ''.join((pre, tab, post))
5278 mobj = get_mobj(url)
5279
5280 # Handle both video/playlist URLs
5281 qs = parse_qs(url)
5282 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5283
5284 if not video_id and mobj['not_channel'].startswith('watch'):
5285 if not playlist_id:
5286 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5287 raise ExtractorError('Unable to recognize tab page')
5288 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5289 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5290 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5291 mobj = get_mobj(url)
5292
5293 if video_id and playlist_id:
5294 if self.get_param('noplaylist'):
5295 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5296 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5297 ie=YoutubeIE.ie_key(), video_id=video_id)
5298 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5299
5300 data, ytcfg = self._extract_data(url, item_id)
5301
5302 # YouTube may provide a non-standard redirect to the regional channel
5303 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5304 redirect_url = traverse_obj(
5305 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5306 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5307 redirect_url = ''.join((
5308 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5309 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5310 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5311
5312 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5313 if tabs:
5314 selected_tab = self._extract_selected_tab(tabs)
5315 selected_tab_name = selected_tab.get('title', '').lower()
5316 if selected_tab_name == 'home':
5317 selected_tab_name = 'featured'
5318 requested_tab_name = mobj['tab'][1:]
5319 if 'no-youtube-channel-redirect' not in compat_opts:
5320 if requested_tab_name == 'live':
5321 # Live tab should have redirected to the video
5322 raise ExtractorError('The channel is not currently live', expected=True)
5323 if requested_tab_name not in ('', selected_tab_name):
5324 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5325 if not original_tab_name:
5326 if item_id[:2] == 'UC':
5327 # Topic channels don't have /videos. Use the equivalent playlist instead
5328 pl_id = f'UU{item_id[2:]}'
5329 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5330 try:
5331 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5332 except ExtractorError:
5333 redirect_warning += ' and the playlist redirect gave error'
5334 else:
5335 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5336 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5337 if selected_tab_name and selected_tab_name != requested_tab_name:
5338 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5339 else:
5340 raise ExtractorError(redirect_warning, expected=True)
5341
5342 if redirect_warning:
5343 self.to_screen(redirect_warning)
5344 self.write_debug(f'Final URL: {url}')
5345
5346 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5347 if 'no-youtube-unavailable-videos' not in compat_opts:
5348 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5349 self._extract_and_report_alerts(data, only_once=True)
5350 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5351 if tabs:
5352 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5353
5354 playlist = traverse_obj(
5355 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5356 if playlist:
5357 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5358
5359 video_id = traverse_obj(
5360 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5361 if video_id:
5362 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5363 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5364 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5365 ie=YoutubeIE.ie_key(), video_id=video_id)
5366
5367 raise ExtractorError('Unable to recognize tab page')
5368
5369
5370 class YoutubePlaylistIE(InfoExtractor):
5371 IE_DESC = 'YouTube playlists'
5372 _VALID_URL = r'''(?x)(?:
5373 (?:https?://)?
5374 (?:\w+\.)?
5375 (?:
5376 (?:
5377 youtube(?:kids)?\.com|
5378 %(invidious)s
5379 )
5380 /.*?\?.*?\blist=
5381 )?
5382 (?P<id>%(playlist_id)s)
5383 )''' % {
5384 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5385 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5386 }
5387 IE_NAME = 'youtube:playlist'
5388 _TESTS = [{
5389 'note': 'issue #673',
5390 'url': 'PLBB231211A4F62143',
5391 'info_dict': {
5392 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5393 'id': 'PLBB231211A4F62143',
5394 'uploader': 'Wickman',
5395 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5396 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5397 'view_count': int,
5398 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5399 'modified_date': r're:\d{8}',
5400 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5401 'channel': 'Wickman',
5402 'tags': [],
5403 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5404 },
5405 'playlist_mincount': 29,
5406 }, {
5407 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5408 'info_dict': {
5409 'title': 'YDL_safe_search',
5410 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5411 },
5412 'playlist_count': 2,
5413 'skip': 'This playlist is private',
5414 }, {
5415 'note': 'embedded',
5416 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5417 'playlist_count': 4,
5418 'info_dict': {
5419 'title': 'JODA15',
5420 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5421 'uploader': 'milan',
5422 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5423 'description': '',
5424 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5425 'tags': [],
5426 'modified_date': '20140919',
5427 'view_count': int,
5428 'channel': 'milan',
5429 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5430 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5431 },
5432 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5433 }, {
5434 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5435 'playlist_mincount': 654,
5436 'info_dict': {
5437 'title': '2018 Chinese New Singles (11/6 updated)',
5438 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5439 'uploader': 'LBK',
5440 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5441 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5442 'channel': 'LBK',
5443 'view_count': int,
5444 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5445 'tags': [],
5446 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5447 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5448 'modified_date': r're:\d{8}',
5449 },
5450 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5451 }, {
5452 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5453 'only_matching': True,
5454 }, {
5455 # music album playlist
5456 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5457 'only_matching': True,
5458 }]
5459
5460 @classmethod
5461 def suitable(cls, url):
5462 if YoutubeTabIE.suitable(url):
5463 return False
5464 from ..utils import parse_qs
5465 qs = parse_qs(url)
5466 if qs.get('v', [None])[0]:
5467 return False
5468 return super().suitable(url)
5469
5470 def _real_extract(self, url):
5471 playlist_id = self._match_id(url)
5472 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5473 url = update_url_query(
5474 'https://www.youtube.com/playlist',
5475 parse_qs(url) or {'list': playlist_id})
5476 if is_music_url:
5477 url = smuggle_url(url, {'is_music_url': True})
5478 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5479
5480
5481 class YoutubeYtBeIE(InfoExtractor):
5482 IE_DESC = 'youtu.be'
5483 _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}
5484 _TESTS = [{
5485 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5486 'info_dict': {
5487 'id': 'yeWKywCrFtk',
5488 'ext': 'mp4',
5489 'title': 'Small Scale Baler and Braiding Rugs',
5490 'uploader': 'Backus-Page House Museum',
5491 'uploader_id': 'backuspagemuseum',
5492 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5493 'upload_date': '20161008',
5494 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5495 'categories': ['Nonprofits & Activism'],
5496 'tags': list,
5497 'like_count': int,
5498 'age_limit': 0,
5499 'playable_in_embed': True,
5500 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5501 'channel': 'Backus-Page House Museum',
5502 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5503 'live_status': 'not_live',
5504 'view_count': int,
5505 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5506 'availability': 'public',
5507 'duration': 59,
5508 },
5509 'params': {
5510 'noplaylist': True,
5511 'skip_download': True,
5512 },
5513 }, {
5514 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5515 'only_matching': True,
5516 }]
5517
5518 def _real_extract(self, url):
5519 mobj = self._match_valid_url(url)
5520 video_id = mobj.group('id')
5521 playlist_id = mobj.group('playlist_id')
5522 return self.url_result(
5523 update_url_query('https://www.youtube.com/watch', {
5524 'v': video_id,
5525 'list': playlist_id,
5526 'feature': 'youtu.be',
5527 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5528
5529
5530 class YoutubeLivestreamEmbedIE(InfoExtractor):
5531 IE_DESC = 'YouTube livestream embeds'
5532 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5533 _TESTS = [{
5534 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5535 'only_matching': True,
5536 }]
5537
5538 def _real_extract(self, url):
5539 channel_id = self._match_id(url)
5540 return self.url_result(
5541 f'https://www.youtube.com/channel/{channel_id}/live',
5542 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5543
5544
5545 class YoutubeYtUserIE(InfoExtractor):
5546 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5547 IE_NAME = 'youtube:user'
5548 _VALID_URL = r'ytuser:(?P<id>.+)'
5549 _TESTS = [{
5550 'url': 'ytuser:phihag',
5551 'only_matching': True,
5552 }]
5553
5554 def _real_extract(self, url):
5555 user_id = self._match_id(url)
5556 return self.url_result(
5557 'https://www.youtube.com/user/%s/videos' % user_id,
5558 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5559
5560
5561 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5562 IE_NAME = 'youtube:favorites'
5563 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5564 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5565 _LOGIN_REQUIRED = True
5566 _TESTS = [{
5567 'url': ':ytfav',
5568 'only_matching': True,
5569 }, {
5570 'url': ':ytfavorites',
5571 'only_matching': True,
5572 }]
5573
5574 def _real_extract(self, url):
5575 return self.url_result(
5576 'https://www.youtube.com/playlist?list=LL',
5577 ie=YoutubeTabIE.ie_key())
5578
5579
5580 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5581 IE_NAME = 'youtube:notif'
5582 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5583 _VALID_URL = r':ytnotif(?:ication)?s?'
5584 _LOGIN_REQUIRED = True
5585 _TESTS = [{
5586 'url': ':ytnotif',
5587 'only_matching': True,
5588 }, {
5589 'url': ':ytnotifications',
5590 'only_matching': True,
5591 }]
5592
5593 def _extract_notification_menu(self, response, continuation_list):
5594 notification_list = traverse_obj(
5595 response,
5596 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5597 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5598 expected_type=list) or []
5599 continuation_list[0] = None
5600 for item in notification_list:
5601 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5602 if entry:
5603 yield entry
5604 continuation = item.get('continuationItemRenderer')
5605 if continuation:
5606 continuation_list[0] = continuation
5607
5608 def _extract_notification_renderer(self, notification):
5609 video_id = traverse_obj(
5610 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5611 url = f'https://www.youtube.com/watch?v={video_id}'
5612 channel_id = None
5613 if not video_id:
5614 browse_ep = traverse_obj(
5615 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5616 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5617 post_id = self._search_regex(
5618 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5619 'post id', default=None)
5620 if not channel_id or not post_id:
5621 return
5622 # The direct /post url redirects to this in the browser
5623 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5624
5625 channel = traverse_obj(
5626 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5627 expected_type=str)
5628 title = self._search_regex(
5629 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5630 'video title', default=None)
5631 if title:
5632 title = title.replace('\xad', '') # remove soft hyphens
5633 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5634 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5635 else None)
5636 return {
5637 '_type': 'url',
5638 'url': url,
5639 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5640 'video_id': video_id,
5641 'title': title,
5642 'channel_id': channel_id,
5643 'channel': channel,
5644 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5645 'upload_date': upload_date,
5646 }
5647
5648 def _notification_menu_entries(self, ytcfg):
5649 continuation_list = [None]
5650 response = None
5651 for page in itertools.count(1):
5652 ctoken = traverse_obj(
5653 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5654 response = self._extract_response(
5655 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5656 ep='notification/get_notification_menu', check_get_keys='actions',
5657 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5658 yield from self._extract_notification_menu(response, continuation_list)
5659 if not continuation_list[0]:
5660 break
5661
5662 def _real_extract(self, url):
5663 display_id = 'notifications'
5664 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5665 self._report_playlist_authcheck(ytcfg)
5666 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5667
5668
5669 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5670 IE_DESC = 'YouTube search'
5671 IE_NAME = 'youtube:search'
5672 _SEARCH_KEY = 'ytsearch'
5673 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5674 _TESTS = [{
5675 'url': 'ytsearch5:youtube-dl test video',
5676 'playlist_count': 5,
5677 'info_dict': {
5678 'id': 'youtube-dl test video',
5679 'title': 'youtube-dl test video',
5680 }
5681 }]
5682
5683
5684 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5685 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5686 _SEARCH_KEY = 'ytsearchdate'
5687 IE_DESC = 'YouTube search, newest videos first'
5688 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5689 _TESTS = [{
5690 'url': 'ytsearchdate5:youtube-dl test video',
5691 'playlist_count': 5,
5692 'info_dict': {
5693 'id': 'youtube-dl test video',
5694 'title': 'youtube-dl test video',
5695 }
5696 }]
5697
5698
5699 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5700 IE_DESC = 'YouTube search URLs with sorting and filter support'
5701 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5702 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5703 _TESTS = [{
5704 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5705 'playlist_mincount': 5,
5706 'info_dict': {
5707 'id': 'youtube-dl test video',
5708 'title': 'youtube-dl test video',
5709 }
5710 }, {
5711 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5712 'playlist_mincount': 5,
5713 'info_dict': {
5714 'id': 'python',
5715 'title': 'python',
5716 }
5717 }, {
5718 'url': 'https://www.youtube.com/results?search_query=%23cats',
5719 'playlist_mincount': 1,
5720 'info_dict': {
5721 'id': '#cats',
5722 'title': '#cats',
5723 'entries': [{
5724 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5725 'title': '#cats',
5726 }],
5727 },
5728 }, {
5729 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5730 'only_matching': True,
5731 }]
5732
5733 def _real_extract(self, url):
5734 qs = parse_qs(url)
5735 query = (qs.get('search_query') or qs.get('q'))[0]
5736 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5737
5738
5739 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5740 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5741 IE_NAME = 'youtube:music:search_url'
5742 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5743 _TESTS = [{
5744 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5745 'playlist_count': 16,
5746 'info_dict': {
5747 'id': 'royalty free music',
5748 'title': 'royalty free music',
5749 }
5750 }, {
5751 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5752 'playlist_mincount': 30,
5753 'info_dict': {
5754 'id': 'royalty free music - songs',
5755 'title': 'royalty free music - songs',
5756 },
5757 'params': {'extract_flat': 'in_playlist'}
5758 }, {
5759 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5760 'playlist_mincount': 30,
5761 'info_dict': {
5762 'id': 'royalty free music - community playlists',
5763 'title': 'royalty free music - community playlists',
5764 },
5765 'params': {'extract_flat': 'in_playlist'}
5766 }]
5767
5768 _SECTIONS = {
5769 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5770 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5771 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5772 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5773 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5774 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5775 }
5776
5777 def _real_extract(self, url):
5778 qs = parse_qs(url)
5779 query = (qs.get('search_query') or qs.get('q'))[0]
5780 params = qs.get('sp', (None,))[0]
5781 if params:
5782 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5783 else:
5784 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5785 params = self._SECTIONS.get(section)
5786 if not params:
5787 section = None
5788 title = join_nonempty(query, section, delim=' - ')
5789 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5790
5791
5792 class YoutubeFeedsInfoExtractor(InfoExtractor):
5793 """
5794 Base class for feed extractors
5795 Subclasses must re-define the _FEED_NAME property.
5796 """
5797 _LOGIN_REQUIRED = True
5798 _FEED_NAME = 'feeds'
5799
5800 def _real_initialize(self):
5801 YoutubeBaseInfoExtractor._check_login_required(self)
5802
5803 @classproperty
5804 def IE_NAME(self):
5805 return f'youtube:{self._FEED_NAME}'
5806
5807 def _real_extract(self, url):
5808 return self.url_result(
5809 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5810
5811
5812 class YoutubeWatchLaterIE(InfoExtractor):
5813 IE_NAME = 'youtube:watchlater'
5814 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5815 _VALID_URL = r':ytwatchlater'
5816 _TESTS = [{
5817 'url': ':ytwatchlater',
5818 'only_matching': True,
5819 }]
5820
5821 def _real_extract(self, url):
5822 return self.url_result(
5823 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5824
5825
5826 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5827 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5828 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5829 _FEED_NAME = 'recommended'
5830 _LOGIN_REQUIRED = False
5831 _TESTS = [{
5832 'url': ':ytrec',
5833 'only_matching': True,
5834 }, {
5835 'url': ':ytrecommended',
5836 'only_matching': True,
5837 }, {
5838 'url': 'https://youtube.com',
5839 'only_matching': True,
5840 }]
5841
5842
5843 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5844 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5845 _VALID_URL = r':ytsub(?:scription)?s?'
5846 _FEED_NAME = 'subscriptions'
5847 _TESTS = [{
5848 'url': ':ytsubs',
5849 'only_matching': True,
5850 }, {
5851 'url': ':ytsubscriptions',
5852 'only_matching': True,
5853 }]
5854
5855
5856 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5857 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5858 _VALID_URL = r':ythis(?:tory)?'
5859 _FEED_NAME = 'history'
5860 _TESTS = [{
5861 'url': ':ythistory',
5862 'only_matching': True,
5863 }]
5864
5865
5866 class YoutubeStoriesIE(InfoExtractor):
5867 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5868 IE_NAME = 'youtube:stories'
5869 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5870 _TESTS = [{
5871 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5872 'only_matching': True,
5873 }]
5874
5875 def _real_extract(self, url):
5876 playlist_id = f'RLTD{self._match_id(url)}'
5877 return self.url_result(
5878 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5879 ie=YoutubeTabIE, video_id=playlist_id)
5880
5881
5882 class YoutubeTruncatedURLIE(InfoExtractor):
5883 IE_NAME = 'youtube:truncated_url'
5884 IE_DESC = False # Do not list
5885 _VALID_URL = r'''(?x)
5886 (?:https?://)?
5887 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5888 (?:watch\?(?:
5889 feature=[a-z_]+|
5890 annotation_id=annotation_[^&]+|
5891 x-yt-cl=[0-9]+|
5892 hl=[^&]*|
5893 t=[0-9]+
5894 )?
5895 |
5896 attribution_link\?a=[^&]+
5897 )
5898 $
5899 '''
5900
5901 _TESTS = [{
5902 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5903 'only_matching': True,
5904 }, {
5905 'url': 'https://www.youtube.com/watch?',
5906 'only_matching': True,
5907 }, {
5908 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5909 'only_matching': True,
5910 }, {
5911 'url': 'https://www.youtube.com/watch?feature=foo',
5912 'only_matching': True,
5913 }, {
5914 'url': 'https://www.youtube.com/watch?hl=en-GB',
5915 'only_matching': True,
5916 }, {
5917 'url': 'https://www.youtube.com/watch?t=2372',
5918 'only_matching': True,
5919 }]
5920
5921 def _real_extract(self, url):
5922 raise ExtractorError(
5923 'Did you forget to quote the URL? Remember that & is a meta '
5924 'character in most shells, so you want to put the URL in quotes, '
5925 'like youtube-dl '
5926 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5927 ' or simply youtube-dl BaW_jenozKc .',
5928 expected=True)
5929
5930
5931 class YoutubeClipIE(InfoExtractor):
5932 IE_NAME = 'youtube:clip'
5933 IE_DESC = False # Do not list
5934 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5935
5936 def _real_extract(self, url):
5937 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5938 return self.url_result(url, 'Generic')
5939
5940
5941 class YoutubeTruncatedIDIE(InfoExtractor):
5942 IE_NAME = 'youtube:truncated_id'
5943 IE_DESC = False # Do not list
5944 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5945
5946 _TESTS = [{
5947 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5948 'only_matching': True,
5949 }]
5950
5951 def _real_extract(self, url):
5952 video_id = self._match_id(url)
5953 raise ExtractorError(
5954 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5955 expected=True)