]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[extractor/youtube] Make signature extraction non-fatal
[yt-dlp.git] / yt_dlp / extractor / youtube.py
... / ...
CommitLineData
1import base64
2import calendar
3import copy
4import datetime
5import hashlib
6import itertools
7import json
8import math
9import os.path
10import random
11import re
12import sys
13import threading
14import time
15import traceback
16
17from .common import InfoExtractor, SearchInfoExtractor
18from ..compat import functools
19from ..compat import (
20 compat_chr,
21 compat_HTTPError,
22 compat_parse_qs,
23 compat_str,
24 compat_urllib_parse_unquote_plus,
25 compat_urllib_parse_urlencode,
26 compat_urllib_parse_urlparse,
27 compat_urlparse,
28)
29from ..jsinterp import JSInterpreter
30from ..utils import (
31 NO_DEFAULT,
32 ExtractorError,
33 bug_reports_message,
34 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
74INNERTUBE_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
233def _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
241def 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
269build_innertube_clients()
270
271
272class 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 and not self._cookies_passed:
398 self.raise_login_required('Login details are needed to download this content', method='cookies')
399
400 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
401 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
402 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
403
404 def _get_default_ytcfg(self, client='web'):
405 return copy.deepcopy(INNERTUBE_CLIENTS[client])
406
407 def _get_innertube_host(self, client='web'):
408 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
409
410 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
411 # try_get but with fallback to default ytcfg client values when present
412 _func = lambda y: try_get(y, getter, expected_type)
413 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
414
415 def _extract_client_name(self, ytcfg, default_client='web'):
416 return self._ytcfg_get_safe(
417 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
418 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
419
420 def _extract_client_version(self, ytcfg, default_client='web'):
421 return self._ytcfg_get_safe(
422 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
423 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
424
425 def _extract_api_key(self, ytcfg=None, default_client='web'):
426 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
427
428 def _extract_context(self, ytcfg=None, default_client='web'):
429 context = get_first(
430 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
431 # Enforce language and tz for extraction
432 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
433 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
434 return context
435
436 _SAPISID = None
437
438 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
439 time_now = round(time.time())
440 if self._SAPISID is None:
441 yt_cookies = self._get_cookies('https://www.youtube.com')
442 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
443 # See: https://github.com/yt-dlp/yt-dlp/issues/393
444 sapisid_cookie = dict_get(
445 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
446 if sapisid_cookie and sapisid_cookie.value:
447 self._SAPISID = sapisid_cookie.value
448 self.write_debug('Extracted SAPISID cookie')
449 # SAPISID cookie is required if not already present
450 if not yt_cookies.get('SAPISID'):
451 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
452 self._set_cookie(
453 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
454 else:
455 self._SAPISID = False
456 if not self._SAPISID:
457 return None
458 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
459 sapisidhash = hashlib.sha1(
460 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
461 return f'SAPISIDHASH {time_now}_{sapisidhash}'
462
463 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
464 note='Downloading API JSON', errnote='Unable to download API page',
465 context=None, api_key=None, api_hostname=None, default_client='web'):
466
467 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
468 data.update(query)
469 real_headers = self.generate_api_headers(default_client=default_client)
470 real_headers.update({'content-type': 'application/json'})
471 if headers:
472 real_headers.update(headers)
473 return self._download_json(
474 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
475 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
476 data=json.dumps(data).encode('utf8'), headers=real_headers,
477 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
478
479 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
480 data = self._search_regex(
481 (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
482 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
483 if data:
484 return self._parse_json(data, item_id, fatal=fatal)
485
486 @staticmethod
487 def _extract_session_index(*data):
488 """
489 Index of current account in account list.
490 See: https://github.com/yt-dlp/yt-dlp/pull/519
491 """
492 for ytcfg in data:
493 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
494 if session_index is not None:
495 return session_index
496
497 # Deprecated?
498 def _extract_identity_token(self, ytcfg=None, webpage=None):
499 if ytcfg:
500 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
501 if token:
502 return token
503 if webpage:
504 return self._search_regex(
505 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
506 'identity token', default=None, fatal=False)
507
508 @staticmethod
509 def _extract_account_syncid(*args):
510 """
511 Extract syncId required to download private playlists of secondary channels
512 @params response and/or ytcfg
513 """
514 for data in args:
515 # ytcfg includes channel_syncid if on secondary channel
516 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
517 if delegated_sid:
518 return delegated_sid
519 sync_ids = (try_get(
520 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
521 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
522 if len(sync_ids) >= 2 and sync_ids[1]:
523 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
524 # and just "user_syncid||" for primary channel. We only want the channel_syncid
525 return sync_ids[0]
526
527 @staticmethod
528 def _extract_visitor_data(*args):
529 """
530 Extracts visitorData from an API response or ytcfg
531 Appears to be used to track session state
532 """
533 return get_first(
534 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
535 expected_type=str)
536
537 @functools.cached_property
538 def is_authenticated(self):
539 return bool(self._generate_sapisidhash_header())
540
541 def extract_ytcfg(self, video_id, webpage):
542 if not webpage:
543 return {}
544 return self._parse_json(
545 self._search_regex(
546 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
547 default='{}'), video_id, fatal=False) or {}
548
549 def generate_api_headers(
550 self, *, ytcfg=None, account_syncid=None, session_index=None,
551 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
552
553 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
554 headers = {
555 'X-YouTube-Client-Name': compat_str(
556 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
557 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
558 'Origin': origin,
559 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
560 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
561 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
562 }
563 if session_index is None:
564 session_index = self._extract_session_index(ytcfg)
565 if account_syncid or session_index is not None:
566 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
567
568 auth = self._generate_sapisidhash_header(origin)
569 if auth is not None:
570 headers['Authorization'] = auth
571 headers['X-Origin'] = origin
572 return {h: v for h, v in headers.items() if v is not None}
573
574 def _download_ytcfg(self, client, video_id):
575 url = {
576 'web': 'https://www.youtube.com',
577 'web_music': 'https://music.youtube.com',
578 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
579 }.get(client)
580 if not url:
581 return {}
582 webpage = self._download_webpage(
583 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
584 return self.extract_ytcfg(video_id, webpage) or {}
585
586 @staticmethod
587 def _build_api_continuation_query(continuation, ctp=None):
588 query = {
589 'continuation': continuation
590 }
591 # TODO: Inconsistency with clickTrackingParams.
592 # Currently we have a fixed ctp contained within context (from ytcfg)
593 # and a ctp in root query for continuation.
594 if ctp:
595 query['clickTracking'] = {'clickTrackingParams': ctp}
596 return query
597
598 @classmethod
599 def _extract_next_continuation_data(cls, renderer):
600 next_continuation = try_get(
601 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
602 lambda x: x['continuation']['reloadContinuationData']), dict)
603 if not next_continuation:
604 return
605 continuation = next_continuation.get('continuation')
606 if not continuation:
607 return
608 ctp = next_continuation.get('clickTrackingParams')
609 return cls._build_api_continuation_query(continuation, ctp)
610
611 @classmethod
612 def _extract_continuation_ep_data(cls, continuation_ep: dict):
613 if isinstance(continuation_ep, dict):
614 continuation = try_get(
615 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
616 if not continuation:
617 return
618 ctp = continuation_ep.get('clickTrackingParams')
619 return cls._build_api_continuation_query(continuation, ctp)
620
621 @classmethod
622 def _extract_continuation(cls, renderer):
623 next_continuation = cls._extract_next_continuation_data(renderer)
624 if next_continuation:
625 return next_continuation
626
627 contents = []
628 for key in ('contents', 'items'):
629 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
630
631 for content in contents:
632 if not isinstance(content, dict):
633 continue
634 continuation_ep = try_get(
635 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
636 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
637 dict)
638 continuation = cls._extract_continuation_ep_data(continuation_ep)
639 if continuation:
640 return continuation
641
642 @classmethod
643 def _extract_alerts(cls, data):
644 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
645 if not isinstance(alert_dict, dict):
646 continue
647 for alert in alert_dict.values():
648 alert_type = alert.get('type')
649 if not alert_type:
650 continue
651 message = cls._get_text(alert, 'text')
652 if message:
653 yield alert_type, message
654
655 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
656 errors = []
657 warnings = []
658 for alert_type, alert_message in alerts:
659 if alert_type.lower() == 'error' and fatal:
660 errors.append([alert_type, alert_message])
661 else:
662 warnings.append([alert_type, alert_message])
663
664 for alert_type, alert_message in (warnings + errors[:-1]):
665 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
666 if errors:
667 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
668
669 def _extract_and_report_alerts(self, data, *args, **kwargs):
670 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
671
672 def _extract_badges(self, renderer: dict):
673 badges = set()
674 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
675 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
676 if label:
677 badges.add(label.lower())
678 return badges
679
680 @staticmethod
681 def _get_text(data, *path_list, max_runs=None):
682 for path in path_list or [None]:
683 if path is None:
684 obj = [data]
685 else:
686 obj = traverse_obj(data, path, default=[])
687 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
688 obj = [obj]
689 for item in obj:
690 text = try_get(item, lambda x: x['simpleText'], compat_str)
691 if text:
692 return text
693 runs = try_get(item, lambda x: x['runs'], list) or []
694 if not runs and isinstance(item, list):
695 runs = item
696
697 runs = runs[:min(len(runs), max_runs or len(runs))]
698 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
699 if text:
700 return text
701
702 def _get_count(self, data, *path_list):
703 count_text = self._get_text(data, *path_list) or ''
704 count = parse_count(count_text)
705 if count is None:
706 count = str_to_int(
707 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
708 return count
709
710 @staticmethod
711 def _extract_thumbnails(data, *path_list):
712 """
713 Extract thumbnails from thumbnails dict
714 @param path_list: path list to level that contains 'thumbnails' key
715 """
716 thumbnails = []
717 for path in path_list or [()]:
718 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
719 thumbnail_url = url_or_none(thumbnail.get('url'))
720 if not thumbnail_url:
721 continue
722 # Sometimes youtube gives a wrong thumbnail URL. See:
723 # https://github.com/yt-dlp/yt-dlp/issues/233
724 # https://github.com/ytdl-org/youtube-dl/issues/28023
725 if 'maxresdefault' in thumbnail_url:
726 thumbnail_url = thumbnail_url.split('?')[0]
727 thumbnails.append({
728 'url': thumbnail_url,
729 'height': int_or_none(thumbnail.get('height')),
730 'width': int_or_none(thumbnail.get('width')),
731 })
732 return thumbnails
733
734 @staticmethod
735 def extract_relative_time(relative_time_text):
736 """
737 Extracts a relative time from string and converts to dt object
738 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
739 """
740 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)
741 if mobj:
742 start = mobj.group('start')
743 if start:
744 return datetime_from_str(start)
745 try:
746 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
747 except ValueError:
748 return None
749
750 def _extract_time_text(self, renderer, *path_list):
751 """@returns (timestamp, time_text)"""
752 text = self._get_text(renderer, *path_list) or ''
753 dt = self.extract_relative_time(text)
754 timestamp = None
755 if isinstance(dt, datetime.datetime):
756 timestamp = calendar.timegm(dt.timetuple())
757
758 if timestamp is None:
759 timestamp = (
760 unified_timestamp(text) or unified_timestamp(
761 self._search_regex(
762 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
763 text.lower(), 'time text', default=None)))
764
765 if text and timestamp is None:
766 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
767 return timestamp, text
768
769 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
770 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
771 default_client='web'):
772 response = None
773 last_error = None
774 count = -1
775 retries = self.get_param('extractor_retries', 3)
776 if check_get_keys is None:
777 check_get_keys = []
778 while count < retries:
779 count += 1
780 if last_error:
781 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
782 try:
783 response = self._call_api(
784 ep=ep, fatal=True, headers=headers,
785 video_id=item_id, query=query,
786 context=self._extract_context(ytcfg, default_client),
787 api_key=self._extract_api_key(ytcfg, default_client),
788 api_hostname=api_hostname, default_client=default_client,
789 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
790 except ExtractorError as e:
791 if isinstance(e.cause, network_exceptions):
792 if isinstance(e.cause, compat_HTTPError):
793 first_bytes = e.cause.read(512)
794 if not is_html(first_bytes):
795 yt_error = try_get(
796 self._parse_json(
797 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
798 lambda x: x['error']['message'], compat_str)
799 if yt_error:
800 self._report_alerts([('ERROR', yt_error)], fatal=False)
801 # Downloading page may result in intermittent 5xx HTTP error
802 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
803 # We also want to catch all other network exceptions since errors in later pages can be troublesome
804 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
805 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
806 last_error = error_to_compat_str(e.cause or e.msg)
807 if count < retries:
808 continue
809 if fatal:
810 raise
811 else:
812 self.report_warning(error_to_compat_str(e))
813 return
814
815 else:
816 try:
817 self._extract_and_report_alerts(response, only_once=True)
818 except ExtractorError as e:
819 # YouTube servers may return errors we want to retry on in a 200 OK response
820 # See: https://github.com/yt-dlp/yt-dlp/issues/839
821 if 'unknown error' in e.msg.lower():
822 last_error = e.msg
823 continue
824 if fatal:
825 raise
826 self.report_warning(error_to_compat_str(e))
827 return
828 if not check_get_keys or dict_get(response, check_get_keys):
829 break
830 # Youtube sometimes sends incomplete data
831 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
832 last_error = 'Incomplete data received'
833 if count >= retries:
834 if fatal:
835 raise ExtractorError(last_error)
836 else:
837 self.report_warning(last_error)
838 return
839 return response
840
841 @staticmethod
842 def is_music_url(url):
843 return re.match(r'https?://music\.youtube\.com/', url) is not None
844
845 def _extract_video(self, renderer):
846 video_id = renderer.get('videoId')
847 title = self._get_text(renderer, 'title')
848 description = self._get_text(renderer, 'descriptionSnippet')
849 duration = parse_duration(self._get_text(
850 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
851 if duration is None:
852 duration = parse_duration(self._search_regex(
853 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
854 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
855 video_id, default=None, group='duration'))
856
857 view_count = self._get_count(renderer, 'viewCountText')
858
859 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
860 channel_id = traverse_obj(
861 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
862 expected_type=str, get_all=False)
863 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
864 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
865 overlay_style = traverse_obj(
866 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
867 get_all=False, expected_type=str)
868 badges = self._extract_badges(renderer)
869 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
870 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
871 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
872 expected_type=str)) or ''
873 url = f'https://www.youtube.com/watch?v={video_id}'
874 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
875 url = f'https://www.youtube.com/shorts/{video_id}'
876
877 return {
878 '_type': 'url',
879 'ie_key': YoutubeIE.ie_key(),
880 'id': video_id,
881 'url': url,
882 'title': title,
883 'description': description,
884 'duration': duration,
885 'view_count': view_count,
886 'uploader': uploader,
887 'channel_id': channel_id,
888 'thumbnails': thumbnails,
889 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
890 if self._configuration_arg('approximate_date', ie_key='youtubetab')
891 else None),
892 'live_status': ('is_upcoming' if scheduled_timestamp is not None
893 else 'was_live' if 'streamed' in time_text.lower()
894 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
895 else None),
896 'release_timestamp': scheduled_timestamp,
897 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
898 }
899
900
901class YoutubeIE(YoutubeBaseInfoExtractor):
902 IE_DESC = 'YouTube'
903 _VALID_URL = r"""(?x)^
904 (
905 (?:https?://|//) # http(s):// or protocol-independent URL
906 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
907 (?:www\.)?deturl\.com/www\.youtube\.com|
908 (?:www\.)?pwnyoutube\.com|
909 (?:www\.)?hooktube\.com|
910 (?:www\.)?yourepeat\.com|
911 tube\.majestyc\.net|
912 %(invidious)s|
913 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
914 (?:.*?\#/)? # handle anchor (#/) redirect urls
915 (?: # the various things that can precede the ID:
916 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
917 |(?: # or the v= param in all its forms
918 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
919 (?:\?|\#!?) # the params delimiter ? or # or #!
920 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
921 v=
922 )
923 ))
924 |(?:
925 youtu\.be| # just youtu.be/xxxx
926 vid\.plus| # or vid.plus/xxxx
927 zwearz\.com/watch| # or zwearz.com/watch/xxxx
928 %(invidious)s
929 )/
930 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
931 )
932 )? # all until now is optional -> you can pass the naked ID
933 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
934 (?(1).+)? # if we found the ID, everything can follow
935 (?:\#|$)""" % {
936 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
937 }
938 _PLAYER_INFO_RE = (
939 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
940 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$',
941 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
942 )
943 _formats = {
944 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
945 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
946 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
947 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
948 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
949 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
950 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
951 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
952 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
953 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
954 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
955 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
956 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
957 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
958 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
959 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
960 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
961 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
962
963
964 # 3D videos
965 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
966 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
967 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
968 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
969 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
970 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
971 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
972
973 # Apple HTTP Live Streaming
974 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
975 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
976 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
977 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
978 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
979 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
980 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
981 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
982
983 # DASH mp4 video
984 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
985 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
986 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
987 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
988 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
989 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
990 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
991 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
992 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
993 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
994 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
995 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
996
997 # Dash mp4 audio
998 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
999 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1000 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1001 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1002 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1003 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1004 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1005
1006 # Dash webm
1007 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1008 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1009 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1010 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1011 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1012 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1013 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1014 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1015 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1016 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1017 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1018 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1019 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1020 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1021 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1022 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1023 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1024 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1025 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1026 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1027 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1028 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1029
1030 # Dash webm audio
1031 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1032 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1033
1034 # Dash webm audio with opus inside
1035 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1036 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1037 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1038
1039 # RTMP (unnamed)
1040 '_rtmp': {'protocol': 'rtmp'},
1041
1042 # av01 video only formats sometimes served with "unknown" codecs
1043 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1044 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1045 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1046 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1047 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1048 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1049 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1050 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1051 }
1052 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1053
1054 _GEO_BYPASS = False
1055
1056 IE_NAME = 'youtube'
1057 _TESTS = [
1058 {
1059 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1060 'info_dict': {
1061 'id': 'BaW_jenozKc',
1062 'ext': 'mp4',
1063 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1064 'uploader': 'Philipp Hagemeister',
1065 'uploader_id': 'phihag',
1066 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1067 'channel': 'Philipp Hagemeister',
1068 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1069 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1070 'upload_date': '20121002',
1071 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1072 'categories': ['Science & Technology'],
1073 'tags': ['youtube-dl'],
1074 'duration': 10,
1075 'view_count': int,
1076 'like_count': int,
1077 'availability': 'public',
1078 'playable_in_embed': True,
1079 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1080 'live_status': 'not_live',
1081 'age_limit': 0,
1082 'start_time': 1,
1083 'end_time': 9,
1084 'channel_follower_count': int
1085 }
1086 },
1087 {
1088 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1089 'note': 'Embed-only video (#1746)',
1090 'info_dict': {
1091 'id': 'yZIXLfi8CZQ',
1092 'ext': 'mp4',
1093 'upload_date': '20120608',
1094 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1095 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1096 'uploader': 'SET India',
1097 'uploader_id': 'setindia',
1098 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1099 'age_limit': 18,
1100 },
1101 'skip': 'Private video',
1102 },
1103 {
1104 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1105 'note': 'Use the first video ID in the URL',
1106 'info_dict': {
1107 'id': 'BaW_jenozKc',
1108 'ext': 'mp4',
1109 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1110 'uploader': 'Philipp Hagemeister',
1111 'uploader_id': 'phihag',
1112 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1113 'channel': 'Philipp Hagemeister',
1114 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1115 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1116 'upload_date': '20121002',
1117 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1118 'categories': ['Science & Technology'],
1119 'tags': ['youtube-dl'],
1120 'duration': 10,
1121 'view_count': int,
1122 'like_count': int,
1123 'availability': 'public',
1124 'playable_in_embed': True,
1125 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1126 'live_status': 'not_live',
1127 'age_limit': 0,
1128 'channel_follower_count': int
1129 },
1130 'params': {
1131 'skip_download': True,
1132 },
1133 },
1134 {
1135 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1136 'note': '256k DASH audio (format 141) via DASH manifest',
1137 'info_dict': {
1138 'id': 'a9LDPn-MO4I',
1139 'ext': 'm4a',
1140 'upload_date': '20121002',
1141 'uploader_id': '8KVIDEO',
1142 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1143 'description': '',
1144 'uploader': '8KVIDEO',
1145 'title': 'UHDTV TEST 8K VIDEO.mp4'
1146 },
1147 'params': {
1148 'youtube_include_dash_manifest': True,
1149 'format': '141',
1150 },
1151 'skip': 'format 141 not served anymore',
1152 },
1153 # DASH manifest with encrypted signature
1154 {
1155 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1156 'info_dict': {
1157 'id': 'IB3lcPjvWLA',
1158 'ext': 'm4a',
1159 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1160 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1161 'duration': 244,
1162 'uploader': 'AfrojackVEVO',
1163 'uploader_id': 'AfrojackVEVO',
1164 'upload_date': '20131011',
1165 'abr': 129.495,
1166 'like_count': int,
1167 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1168 'playable_in_embed': True,
1169 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1170 'view_count': int,
1171 'track': 'The Spark',
1172 'live_status': 'not_live',
1173 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1174 'channel': 'Afrojack',
1175 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1176 'tags': 'count:19',
1177 'availability': 'public',
1178 'categories': ['Music'],
1179 'age_limit': 0,
1180 'alt_title': 'The Spark',
1181 'channel_follower_count': int
1182 },
1183 'params': {
1184 'youtube_include_dash_manifest': True,
1185 'format': '141/bestaudio[ext=m4a]',
1186 },
1187 },
1188 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1189 {
1190 'note': 'Embed allowed age-gate video',
1191 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1192 'info_dict': {
1193 'id': 'HtVdAasjOgU',
1194 'ext': 'mp4',
1195 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1196 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1197 'duration': 142,
1198 'uploader': 'The Witcher',
1199 'uploader_id': 'WitcherGame',
1200 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1201 'upload_date': '20140605',
1202 'age_limit': 18,
1203 'categories': ['Gaming'],
1204 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1205 'availability': 'needs_auth',
1206 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1207 'like_count': int,
1208 'channel': 'The Witcher',
1209 'live_status': 'not_live',
1210 'tags': 'count:17',
1211 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1212 'playable_in_embed': True,
1213 'view_count': int,
1214 'channel_follower_count': int
1215 },
1216 },
1217 {
1218 'note': 'Age-gate video with embed allowed in public site',
1219 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1220 'info_dict': {
1221 'id': 'HsUATh_Nc2U',
1222 'ext': 'mp4',
1223 'title': 'Godzilla 2 (Official Video)',
1224 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1225 'upload_date': '20200408',
1226 'uploader_id': 'FlyingKitty900',
1227 'uploader': 'FlyingKitty',
1228 'age_limit': 18,
1229 'availability': 'needs_auth',
1230 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1231 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1232 'channel': 'FlyingKitty',
1233 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1234 'view_count': int,
1235 'categories': ['Entertainment'],
1236 'live_status': 'not_live',
1237 'tags': ['Flyingkitty', 'godzilla 2'],
1238 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1239 'like_count': int,
1240 'duration': 177,
1241 'playable_in_embed': True,
1242 'channel_follower_count': int
1243 },
1244 },
1245 {
1246 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1247 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1248 'info_dict': {
1249 'id': 'Tq92D6wQ1mg',
1250 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1251 'ext': 'mp4',
1252 'upload_date': '20191228',
1253 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1254 'uploader': 'Projekt Melody',
1255 'description': 'md5:17eccca93a786d51bc67646756894066',
1256 'age_limit': 18,
1257 'like_count': int,
1258 'availability': 'needs_auth',
1259 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1260 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1261 'view_count': int,
1262 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1263 'channel': 'Projekt Melody',
1264 'live_status': 'not_live',
1265 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1266 'playable_in_embed': True,
1267 'categories': ['Entertainment'],
1268 'duration': 106,
1269 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1270 'channel_follower_count': int
1271 },
1272 },
1273 {
1274 'note': 'Non-Agegated non-embeddable video',
1275 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1276 'info_dict': {
1277 'id': 'MeJVWBSsPAY',
1278 'ext': 'mp4',
1279 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1280 'uploader': 'Herr Lurik',
1281 'uploader_id': 'st3in234',
1282 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1283 'upload_date': '20130730',
1284 'track': 'Such mich find mich',
1285 'age_limit': 0,
1286 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1287 'like_count': int,
1288 'playable_in_embed': False,
1289 'creator': 'OOMPH!',
1290 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1291 'view_count': int,
1292 'alt_title': 'Such mich find mich',
1293 'duration': 210,
1294 'channel': 'Herr Lurik',
1295 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1296 'categories': ['Music'],
1297 'availability': 'public',
1298 'uploader_url': 'http://www.youtube.com/user/st3in234',
1299 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1300 'live_status': 'not_live',
1301 'artist': 'OOMPH!',
1302 'channel_follower_count': int
1303 },
1304 },
1305 {
1306 'note': 'Non-bypassable age-gated video',
1307 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1308 'only_matching': True,
1309 },
1310 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1311 # YouTube Red ad is not captured for creator
1312 {
1313 'url': '__2ABJjxzNo',
1314 'info_dict': {
1315 'id': '__2ABJjxzNo',
1316 'ext': 'mp4',
1317 'duration': 266,
1318 'upload_date': '20100430',
1319 'uploader_id': 'deadmau5',
1320 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1321 'creator': 'deadmau5',
1322 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1323 'uploader': 'deadmau5',
1324 'title': 'Deadmau5 - Some Chords (HD)',
1325 'alt_title': 'Some Chords',
1326 'availability': 'public',
1327 'tags': 'count:14',
1328 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1329 'view_count': int,
1330 'live_status': 'not_live',
1331 'channel': 'deadmau5',
1332 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1333 'like_count': int,
1334 'track': 'Some Chords',
1335 'artist': 'deadmau5',
1336 'playable_in_embed': True,
1337 'age_limit': 0,
1338 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1339 'categories': ['Music'],
1340 'album': 'Some Chords',
1341 'channel_follower_count': int
1342 },
1343 'expected_warnings': [
1344 'DASH manifest missing',
1345 ]
1346 },
1347 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1348 {
1349 'url': 'lqQg6PlCWgI',
1350 'info_dict': {
1351 'id': 'lqQg6PlCWgI',
1352 'ext': 'mp4',
1353 'duration': 6085,
1354 'upload_date': '20150827',
1355 'uploader_id': 'olympic',
1356 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1357 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1358 'uploader': 'Olympics',
1359 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1360 'like_count': int,
1361 'release_timestamp': 1343767800,
1362 'playable_in_embed': True,
1363 'categories': ['Sports'],
1364 'release_date': '20120731',
1365 'channel': 'Olympics',
1366 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1367 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1368 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1369 'age_limit': 0,
1370 'availability': 'public',
1371 'live_status': 'was_live',
1372 'view_count': int,
1373 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1374 'channel_follower_count': int
1375 },
1376 'params': {
1377 'skip_download': 'requires avconv',
1378 }
1379 },
1380 # Non-square pixels
1381 {
1382 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1383 'info_dict': {
1384 'id': '_b-2C3KPAM0',
1385 'ext': 'mp4',
1386 'stretched_ratio': 16 / 9.,
1387 'duration': 85,
1388 'upload_date': '20110310',
1389 'uploader_id': 'AllenMeow',
1390 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1391 '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',
1392 'uploader': '孫ᄋᄅ',
1393 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1394 'playable_in_embed': True,
1395 'channel': '孫ᄋᄅ',
1396 'age_limit': 0,
1397 'tags': 'count:11',
1398 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1399 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1400 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1401 'view_count': int,
1402 'categories': ['People & Blogs'],
1403 'like_count': int,
1404 'live_status': 'not_live',
1405 'availability': 'unlisted',
1406 'channel_follower_count': int
1407 },
1408 },
1409 # url_encoded_fmt_stream_map is empty string
1410 {
1411 'url': 'qEJwOuvDf7I',
1412 'info_dict': {
1413 'id': 'qEJwOuvDf7I',
1414 'ext': 'webm',
1415 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1416 'description': '',
1417 'upload_date': '20150404',
1418 'uploader_id': 'spbelect',
1419 'uploader': 'Наблюдатели Петербурга',
1420 },
1421 'params': {
1422 'skip_download': 'requires avconv',
1423 },
1424 'skip': 'This live event has ended.',
1425 },
1426 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1427 {
1428 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1429 'info_dict': {
1430 'id': 'FIl7x6_3R5Y',
1431 'ext': 'webm',
1432 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1433 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1434 'duration': 220,
1435 'upload_date': '20150625',
1436 'uploader_id': 'dorappi2000',
1437 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1438 'uploader': 'dorappi2000',
1439 'formats': 'mincount:31',
1440 },
1441 'skip': 'not actual anymore',
1442 },
1443 # DASH manifest with segment_list
1444 {
1445 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1446 'md5': '8ce563a1d667b599d21064e982ab9e31',
1447 'info_dict': {
1448 'id': 'CsmdDsKjzN8',
1449 'ext': 'mp4',
1450 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1451 'uploader': 'Airtek',
1452 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1453 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1454 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1455 },
1456 'params': {
1457 'youtube_include_dash_manifest': True,
1458 'format': '135', # bestvideo
1459 },
1460 'skip': 'This live event has ended.',
1461 },
1462 {
1463 # Multifeed videos (multiple cameras), URL is for Main Camera
1464 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1465 'info_dict': {
1466 'id': 'jvGDaLqkpTg',
1467 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1468 'description': 'md5:e03b909557865076822aa169218d6a5d',
1469 },
1470 'playlist': [{
1471 'info_dict': {
1472 'id': 'jvGDaLqkpTg',
1473 'ext': 'mp4',
1474 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1475 'description': 'md5:e03b909557865076822aa169218d6a5d',
1476 'duration': 10643,
1477 'upload_date': '20161111',
1478 'uploader': 'Team PGP',
1479 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1480 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1481 },
1482 }, {
1483 'info_dict': {
1484 'id': '3AKt1R1aDnw',
1485 'ext': 'mp4',
1486 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1487 'description': 'md5:e03b909557865076822aa169218d6a5d',
1488 'duration': 10991,
1489 'upload_date': '20161111',
1490 'uploader': 'Team PGP',
1491 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1492 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1493 },
1494 }, {
1495 'info_dict': {
1496 'id': 'RtAMM00gpVc',
1497 'ext': 'mp4',
1498 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1499 'description': 'md5:e03b909557865076822aa169218d6a5d',
1500 'duration': 10995,
1501 'upload_date': '20161111',
1502 'uploader': 'Team PGP',
1503 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1504 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1505 },
1506 }, {
1507 'info_dict': {
1508 'id': '6N2fdlP3C5U',
1509 'ext': 'mp4',
1510 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1511 'description': 'md5:e03b909557865076822aa169218d6a5d',
1512 'duration': 10990,
1513 'upload_date': '20161111',
1514 'uploader': 'Team PGP',
1515 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1516 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1517 },
1518 }],
1519 'params': {
1520 'skip_download': True,
1521 },
1522 'skip': 'Not multifeed anymore',
1523 },
1524 {
1525 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1526 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1527 'info_dict': {
1528 'id': 'gVfLd0zydlo',
1529 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1530 },
1531 'playlist_count': 2,
1532 'skip': 'Not multifeed anymore',
1533 },
1534 {
1535 'url': 'https://vid.plus/FlRa-iH7PGw',
1536 'only_matching': True,
1537 },
1538 {
1539 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1540 'only_matching': True,
1541 },
1542 {
1543 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1544 # Also tests cut-off URL expansion in video description (see
1545 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1546 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1547 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1548 'info_dict': {
1549 'id': 'lsguqyKfVQg',
1550 'ext': 'mp4',
1551 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1552 'alt_title': 'Dark Walk',
1553 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1554 'duration': 133,
1555 'upload_date': '20151119',
1556 'uploader_id': 'IronSoulElf',
1557 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1558 'uploader': 'IronSoulElf',
1559 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1560 'track': 'Dark Walk',
1561 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1562 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1563 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1564 'categories': ['Film & Animation'],
1565 'view_count': int,
1566 'live_status': 'not_live',
1567 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1568 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1569 'tags': 'count:13',
1570 'availability': 'public',
1571 'channel': 'IronSoulElf',
1572 'playable_in_embed': True,
1573 'like_count': int,
1574 'age_limit': 0,
1575 'channel_follower_count': int
1576 },
1577 'params': {
1578 'skip_download': True,
1579 },
1580 },
1581 {
1582 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1583 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1584 'only_matching': True,
1585 },
1586 {
1587 # Video with yt:stretch=17:0
1588 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1589 'info_dict': {
1590 'id': 'Q39EVAstoRM',
1591 'ext': 'mp4',
1592 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1593 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1594 'upload_date': '20151107',
1595 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1596 'uploader': 'CH GAMER DROID',
1597 },
1598 'params': {
1599 'skip_download': True,
1600 },
1601 'skip': 'This video does not exist.',
1602 },
1603 {
1604 # Video with incomplete 'yt:stretch=16:'
1605 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1606 'only_matching': True,
1607 },
1608 {
1609 # Video licensed under Creative Commons
1610 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1611 'info_dict': {
1612 'id': 'M4gD1WSo5mA',
1613 'ext': 'mp4',
1614 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1615 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1616 'duration': 721,
1617 'upload_date': '20150128',
1618 'uploader_id': 'BerkmanCenter',
1619 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1620 'uploader': 'The Berkman Klein Center for Internet & Society',
1621 'license': 'Creative Commons Attribution license (reuse allowed)',
1622 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1623 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1624 'like_count': int,
1625 'age_limit': 0,
1626 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1627 'channel': 'The Berkman Klein Center for Internet & Society',
1628 'availability': 'public',
1629 'view_count': int,
1630 'categories': ['Education'],
1631 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1632 'live_status': 'not_live',
1633 'playable_in_embed': True,
1634 'channel_follower_count': int
1635 },
1636 'params': {
1637 'skip_download': True,
1638 },
1639 },
1640 {
1641 # Channel-like uploader_url
1642 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1643 'info_dict': {
1644 'id': 'eQcmzGIKrzg',
1645 'ext': 'mp4',
1646 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1647 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1648 'duration': 4060,
1649 'upload_date': '20151120',
1650 'uploader': 'Bernie Sanders',
1651 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1652 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1653 'license': 'Creative Commons Attribution license (reuse allowed)',
1654 'playable_in_embed': True,
1655 'tags': 'count:12',
1656 'like_count': int,
1657 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1658 'age_limit': 0,
1659 'availability': 'public',
1660 'categories': ['News & Politics'],
1661 'channel': 'Bernie Sanders',
1662 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1663 'view_count': int,
1664 'live_status': 'not_live',
1665 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1666 'channel_follower_count': int
1667 },
1668 'params': {
1669 'skip_download': True,
1670 },
1671 },
1672 {
1673 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1674 'only_matching': True,
1675 },
1676 {
1677 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1678 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1679 'only_matching': True,
1680 },
1681 {
1682 # Rental video preview
1683 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1684 'info_dict': {
1685 'id': 'uGpuVWrhIzE',
1686 'ext': 'mp4',
1687 'title': 'Piku - Trailer',
1688 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1689 'upload_date': '20150811',
1690 'uploader': 'FlixMatrix',
1691 'uploader_id': 'FlixMatrixKaravan',
1692 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1693 'license': 'Standard YouTube License',
1694 },
1695 'params': {
1696 'skip_download': True,
1697 },
1698 'skip': 'This video is not available.',
1699 },
1700 {
1701 # YouTube Red video with episode data
1702 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1703 'info_dict': {
1704 'id': 'iqKdEhx-dD4',
1705 'ext': 'mp4',
1706 'title': 'Isolation - Mind Field (Ep 1)',
1707 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1708 'duration': 2085,
1709 'upload_date': '20170118',
1710 'uploader': 'Vsauce',
1711 'uploader_id': 'Vsauce',
1712 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1713 'series': 'Mind Field',
1714 'season_number': 1,
1715 'episode_number': 1,
1716 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1717 'tags': 'count:12',
1718 'view_count': int,
1719 'availability': 'public',
1720 'age_limit': 0,
1721 'channel': 'Vsauce',
1722 'episode': 'Episode 1',
1723 'categories': ['Entertainment'],
1724 'season': 'Season 1',
1725 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1726 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1727 'like_count': int,
1728 'playable_in_embed': True,
1729 'live_status': 'not_live',
1730 'channel_follower_count': int
1731 },
1732 'params': {
1733 'skip_download': True,
1734 },
1735 'expected_warnings': [
1736 'Skipping DASH manifest',
1737 ],
1738 },
1739 {
1740 # The following content has been identified by the YouTube community
1741 # as inappropriate or offensive to some audiences.
1742 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1743 'info_dict': {
1744 'id': '6SJNVb0GnPI',
1745 'ext': 'mp4',
1746 'title': 'Race Differences in Intelligence',
1747 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1748 'duration': 965,
1749 'upload_date': '20140124',
1750 'uploader': 'New Century Foundation',
1751 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1752 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1753 },
1754 'params': {
1755 'skip_download': True,
1756 },
1757 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1758 },
1759 {
1760 # itag 212
1761 'url': '1t24XAntNCY',
1762 'only_matching': True,
1763 },
1764 {
1765 # geo restricted to JP
1766 'url': 'sJL6WA-aGkQ',
1767 'only_matching': True,
1768 },
1769 {
1770 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1771 'only_matching': True,
1772 },
1773 {
1774 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1775 'only_matching': True,
1776 },
1777 {
1778 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1779 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1780 'only_matching': True,
1781 },
1782 {
1783 # DRM protected
1784 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1785 'only_matching': True,
1786 },
1787 {
1788 # Video with unsupported adaptive stream type formats
1789 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1790 'info_dict': {
1791 'id': 'Z4Vy8R84T1U',
1792 'ext': 'mp4',
1793 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1794 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1795 'duration': 433,
1796 'upload_date': '20130923',
1797 'uploader': 'Amelia Putri Harwita',
1798 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1799 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1800 'formats': 'maxcount:10',
1801 },
1802 'params': {
1803 'skip_download': True,
1804 'youtube_include_dash_manifest': False,
1805 },
1806 'skip': 'not actual anymore',
1807 },
1808 {
1809 # Youtube Music Auto-generated description
1810 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1811 'info_dict': {
1812 'id': 'MgNrAu2pzNs',
1813 'ext': 'mp4',
1814 'title': 'Voyeur Girl',
1815 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1816 'upload_date': '20190312',
1817 'uploader': 'Stephen - Topic',
1818 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1819 'artist': 'Stephen',
1820 'track': 'Voyeur Girl',
1821 'album': 'it\'s too much love to know my dear',
1822 'release_date': '20190313',
1823 'release_year': 2019,
1824 'alt_title': 'Voyeur Girl',
1825 'view_count': int,
1826 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1827 'playable_in_embed': True,
1828 'like_count': int,
1829 'categories': ['Music'],
1830 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1831 'channel': 'Stephen',
1832 'availability': 'public',
1833 'creator': 'Stephen',
1834 'duration': 169,
1835 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1836 'age_limit': 0,
1837 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1838 'tags': 'count:11',
1839 'live_status': 'not_live',
1840 'channel_follower_count': int
1841 },
1842 'params': {
1843 'skip_download': True,
1844 },
1845 },
1846 {
1847 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1848 'only_matching': True,
1849 },
1850 {
1851 # invalid -> valid video id redirection
1852 'url': 'DJztXj2GPfl',
1853 'info_dict': {
1854 'id': 'DJztXj2GPfk',
1855 'ext': 'mp4',
1856 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1857 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1858 'upload_date': '20090125',
1859 'uploader': 'Prochorowka',
1860 'uploader_id': 'Prochorowka',
1861 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1862 'artist': 'Panjabi MC',
1863 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1864 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1865 },
1866 'params': {
1867 'skip_download': True,
1868 },
1869 'skip': 'Video unavailable',
1870 },
1871 {
1872 # empty description results in an empty string
1873 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1874 'info_dict': {
1875 'id': 'x41yOUIvK2k',
1876 'ext': 'mp4',
1877 'title': 'IMG 3456',
1878 'description': '',
1879 'upload_date': '20170613',
1880 'uploader_id': 'ElevageOrVert',
1881 'uploader': 'ElevageOrVert',
1882 'view_count': int,
1883 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1884 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1885 'like_count': int,
1886 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1887 'tags': [],
1888 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1889 'availability': 'public',
1890 'age_limit': 0,
1891 'categories': ['Pets & Animals'],
1892 'duration': 7,
1893 'playable_in_embed': True,
1894 'live_status': 'not_live',
1895 'channel': 'ElevageOrVert',
1896 'channel_follower_count': int
1897 },
1898 'params': {
1899 'skip_download': True,
1900 },
1901 },
1902 {
1903 # with '};' inside yt initial data (see [1])
1904 # see [2] for an example with '};' inside ytInitialPlayerResponse
1905 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1906 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1907 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1908 'info_dict': {
1909 'id': 'CHqg6qOn4no',
1910 'ext': 'mp4',
1911 'title': 'Part 77 Sort a list of simple types in c#',
1912 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1913 'upload_date': '20130831',
1914 'uploader_id': 'kudvenkat',
1915 'uploader': 'kudvenkat',
1916 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1917 'like_count': int,
1918 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1919 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1920 'live_status': 'not_live',
1921 'categories': ['Education'],
1922 'availability': 'public',
1923 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1924 'tags': 'count:12',
1925 'playable_in_embed': True,
1926 'age_limit': 0,
1927 'view_count': int,
1928 'duration': 522,
1929 'channel': 'kudvenkat',
1930 'channel_follower_count': int
1931 },
1932 'params': {
1933 'skip_download': True,
1934 },
1935 },
1936 {
1937 # another example of '};' in ytInitialData
1938 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1939 'only_matching': True,
1940 },
1941 {
1942 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1943 'only_matching': True,
1944 },
1945 {
1946 # https://github.com/ytdl-org/youtube-dl/pull/28094
1947 'url': 'OtqTfy26tG0',
1948 'info_dict': {
1949 'id': 'OtqTfy26tG0',
1950 'ext': 'mp4',
1951 'title': 'Burn Out',
1952 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1953 'upload_date': '20141120',
1954 'uploader': 'The Cinematic Orchestra - Topic',
1955 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1956 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1957 'artist': 'The Cinematic Orchestra',
1958 'track': 'Burn Out',
1959 'album': 'Every Day',
1960 'like_count': int,
1961 'live_status': 'not_live',
1962 'alt_title': 'Burn Out',
1963 'duration': 614,
1964 'age_limit': 0,
1965 'view_count': int,
1966 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1967 'creator': 'The Cinematic Orchestra',
1968 'channel': 'The Cinematic Orchestra',
1969 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1970 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1971 'availability': 'public',
1972 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1973 'categories': ['Music'],
1974 'playable_in_embed': True,
1975 'channel_follower_count': int
1976 },
1977 'params': {
1978 'skip_download': True,
1979 },
1980 },
1981 {
1982 # controversial video, only works with bpctr when authenticated with cookies
1983 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1984 'only_matching': True,
1985 },
1986 {
1987 # controversial video, requires bpctr/contentCheckOk
1988 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1989 'info_dict': {
1990 'id': 'SZJvDhaSDnc',
1991 'ext': 'mp4',
1992 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1993 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1994 'uploader': 'CBS Mornings',
1995 'uploader_id': 'CBSThisMorning',
1996 'upload_date': '20140716',
1997 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
1998 'duration': 170,
1999 'categories': ['News & Politics'],
2000 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2001 'view_count': int,
2002 'channel': 'CBS Mornings',
2003 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2004 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2005 'age_limit': 18,
2006 'availability': 'needs_auth',
2007 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2008 'like_count': int,
2009 'live_status': 'not_live',
2010 'playable_in_embed': True,
2011 'channel_follower_count': int
2012 }
2013 },
2014 {
2015 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2016 'url': 'cBvYw8_A0vQ',
2017 'info_dict': {
2018 'id': 'cBvYw8_A0vQ',
2019 'ext': 'mp4',
2020 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2021 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2022 'upload_date': '20201120',
2023 'uploader': 'Walk around Japan',
2024 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2026 'duration': 1456,
2027 'categories': ['Travel & Events'],
2028 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2029 'view_count': int,
2030 'channel': 'Walk around Japan',
2031 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2032 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2033 'age_limit': 0,
2034 'availability': 'public',
2035 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2036 'live_status': 'not_live',
2037 'playable_in_embed': True,
2038 'channel_follower_count': int
2039 },
2040 'params': {
2041 'skip_download': True,
2042 },
2043 }, {
2044 # Has multiple audio streams
2045 'url': 'WaOKSUlf4TM',
2046 'only_matching': True
2047 }, {
2048 # Requires Premium: has format 141 when requested using YTM url
2049 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2050 'only_matching': True
2051 }, {
2052 # multiple subtitles with same lang_code
2053 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2054 'only_matching': True,
2055 }, {
2056 # Force use android client fallback
2057 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2058 'info_dict': {
2059 'id': 'YOelRv7fMxY',
2060 'title': 'DIGGING A SECRET TUNNEL Part 1',
2061 'ext': '3gp',
2062 'upload_date': '20210624',
2063 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2064 'uploader': 'colinfurze',
2065 'uploader_id': 'colinfurze',
2066 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2067 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2068 'duration': 596,
2069 'categories': ['Entertainment'],
2070 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2071 'view_count': int,
2072 'channel': 'colinfurze',
2073 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2074 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2075 'age_limit': 0,
2076 'availability': 'public',
2077 'like_count': int,
2078 'live_status': 'not_live',
2079 'playable_in_embed': True,
2080 'channel_follower_count': int
2081 },
2082 'params': {
2083 'format': '17', # 3gp format available on android
2084 'extractor_args': {'youtube': {'player_client': ['android']}},
2085 },
2086 },
2087 {
2088 # Skip download of additional client configs (remix client config in this case)
2089 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2090 'only_matching': True,
2091 'params': {
2092 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2093 },
2094 }, {
2095 # shorts
2096 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2097 'only_matching': True,
2098 }, {
2099 'note': 'Storyboards',
2100 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2101 'info_dict': {
2102 'id': '5KLPxDtMqe8',
2103 'ext': 'mhtml',
2104 'format_id': 'sb0',
2105 'title': 'Your Brain is Plastic',
2106 'uploader_id': 'scishow',
2107 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2108 'upload_date': '20140324',
2109 'uploader': 'SciShow',
2110 'like_count': int,
2111 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2112 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2113 'view_count': int,
2114 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2115 'playable_in_embed': True,
2116 'tags': 'count:12',
2117 'uploader_url': 'http://www.youtube.com/user/scishow',
2118 'availability': 'public',
2119 'channel': 'SciShow',
2120 'live_status': 'not_live',
2121 'duration': 248,
2122 'categories': ['Education'],
2123 'age_limit': 0,
2124 'channel_follower_count': int
2125 }, 'params': {'format': 'mhtml', 'skip_download': True}
2126 }, {
2127 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2128 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2129 'info_dict': {
2130 'id': '2NUZ8W2llS4',
2131 'ext': 'mp4',
2132 'title': 'The NP that test your phone performance 🙂',
2133 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2134 'uploader': 'Leon Nguyen',
2135 'uploader_id': 'VNSXIII',
2136 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2137 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2138 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2139 'duration': 21,
2140 'view_count': int,
2141 'age_limit': 0,
2142 'categories': ['Gaming'],
2143 'tags': 'count:23',
2144 'playable_in_embed': True,
2145 'live_status': 'not_live',
2146 'upload_date': '20220103',
2147 'like_count': int,
2148 'availability': 'public',
2149 'channel': 'Leon Nguyen',
2150 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2151 'channel_follower_count': int
2152 }
2153 }, {
2154 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2155 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2156 'info_dict': {
2157 'id': 'mzZzzBU6lrM',
2158 'ext': 'mp4',
2159 'title': 'I Met GeorgeNotFound In Real Life...',
2160 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2161 'uploader': 'Quackity',
2162 'uploader_id': 'QuackityHQ',
2163 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2164 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2165 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2166 'duration': 955,
2167 'view_count': int,
2168 'age_limit': 0,
2169 'categories': ['Entertainment'],
2170 'tags': 'count:26',
2171 'playable_in_embed': True,
2172 'live_status': 'not_live',
2173 'release_timestamp': 1641172509,
2174 'release_date': '20220103',
2175 'upload_date': '20220103',
2176 'like_count': int,
2177 'availability': 'public',
2178 'channel': 'Quackity',
2179 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2180 'channel_follower_count': int
2181 }
2182 },
2183 { # continuous livestream. Microformat upload date should be preferred.
2184 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2185 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2186 'info_dict': {
2187 'id': 'kgx4WGK0oNU',
2188 '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}',
2189 'ext': 'mp4',
2190 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2191 'availability': 'public',
2192 'age_limit': 0,
2193 'release_timestamp': 1637975704,
2194 'upload_date': '20210619',
2195 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2196 'live_status': 'is_live',
2197 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2198 'uploader': '阿鲍Abao',
2199 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2200 'channel': 'Abao in Tokyo',
2201 'channel_follower_count': int,
2202 'release_date': '20211127',
2203 'tags': 'count:39',
2204 'categories': ['People & Blogs'],
2205 'like_count': int,
2206 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2207 'view_count': int,
2208 'playable_in_embed': True,
2209 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2210 },
2211 'params': {'skip_download': True}
2212 }, {
2213 # Story. Requires specific player params to work.
2214 # Note: stories get removed after some period of time
2215 'url': 'https://www.youtube.com/watch?v=yN3x1t3sieA',
2216 'info_dict': {
2217 'id': 'yN3x1t3sieA',
2218 'ext': 'mp4',
2219 'uploader': 'Linus Tech Tips',
2220 'duration': 13,
2221 'channel': 'Linus Tech Tips',
2222 'playable_in_embed': True,
2223 'tags': [],
2224 'age_limit': 0,
2225 'uploader_url': 'http://www.youtube.com/user/LinusTechTips',
2226 'upload_date': '20220402',
2227 'thumbnail': 'https://i.ytimg.com/vi_webp/yN3x1t3sieA/maxresdefault.webp',
2228 'title': 'Story',
2229 'live_status': 'not_live',
2230 'uploader_id': 'LinusTechTips',
2231 'view_count': int,
2232 'description': '',
2233 'channel_id': 'UCXuqSBlHAE6Xw-yeJA0Tunw',
2234 'categories': ['Science & Technology'],
2235 'channel_url': 'https://www.youtube.com/channel/UCXuqSBlHAE6Xw-yeJA0Tunw',
2236 'availability': 'unlisted',
2237 }
2238 }
2239 ]
2240
2241 @classmethod
2242 def suitable(cls, url):
2243 from ..utils import parse_qs
2244
2245 qs = parse_qs(url)
2246 if qs.get('list', [None])[0]:
2247 return False
2248 return super().suitable(url)
2249
2250 def __init__(self, *args, **kwargs):
2251 super().__init__(*args, **kwargs)
2252 self._code_cache = {}
2253 self._player_cache = {}
2254
2255 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2256 lock = threading.Lock()
2257
2258 is_live = True
2259 start_time = time.time()
2260 formats = [f for f in formats if f.get('is_from_start')]
2261
2262 def refetch_manifest(format_id, delay):
2263 nonlocal formats, start_time, is_live
2264 if time.time() <= start_time + delay:
2265 return
2266
2267 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2268 video_details = traverse_obj(
2269 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2270 microformats = traverse_obj(
2271 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2272 expected_type=dict, default=[])
2273 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2274 start_time = time.time()
2275
2276 def mpd_feed(format_id, delay):
2277 """
2278 @returns (manifest_url, manifest_stream_number, is_live) or None
2279 """
2280 with lock:
2281 refetch_manifest(format_id, delay)
2282
2283 f = next((f for f in formats if f['format_id'] == format_id), None)
2284 if not f:
2285 if not is_live:
2286 self.to_screen(f'{video_id}: Video is no longer live')
2287 else:
2288 self.report_warning(
2289 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2290 return None
2291 return f['manifest_url'], f['manifest_stream_number'], is_live
2292
2293 for f in formats:
2294 f['is_live'] = True
2295 f['protocol'] = 'http_dash_segments_generator'
2296 f['fragments'] = functools.partial(
2297 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2298
2299 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2300 FETCH_SPAN, MAX_DURATION = 5, 432000
2301
2302 mpd_url, stream_number, is_live = None, None, True
2303
2304 begin_index = 0
2305 download_start_time = ctx.get('start') or time.time()
2306
2307 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2308 if lack_early_segments:
2309 self.report_warning(bug_reports_message(
2310 'Starting download from the last 120 hours of the live stream since '
2311 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2312 lack_early_segments = True
2313
2314 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2315 fragments, fragment_base_url = None, None
2316
2317 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2318 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2319 # Obtain from MPD's maximum seq value
2320 old_mpd_url = mpd_url
2321 last_error = ctx.pop('last_error', None)
2322 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2323 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2324 or (mpd_url, stream_number, False))
2325 if not refresh_sequence:
2326 if expire_fast and not is_live:
2327 return False, last_seq
2328 elif old_mpd_url == mpd_url:
2329 return True, last_seq
2330 try:
2331 fmts, _ = self._extract_mpd_formats_and_subtitles(
2332 mpd_url, None, note=False, errnote=False, fatal=False)
2333 except ExtractorError:
2334 fmts = None
2335 if not fmts:
2336 no_fragment_score += 2
2337 return False, last_seq
2338 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2339 fragments = fmt_info['fragments']
2340 fragment_base_url = fmt_info['fragment_base_url']
2341 assert fragment_base_url
2342
2343 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2344 return True, _last_seq
2345
2346 while is_live:
2347 fetch_time = time.time()
2348 if no_fragment_score > 30:
2349 return
2350 if last_segment_url:
2351 # Obtain from "X-Head-Seqnum" header value from each segment
2352 try:
2353 urlh = self._request_webpage(
2354 last_segment_url, None, note=False, errnote=False, fatal=False)
2355 except ExtractorError:
2356 urlh = None
2357 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2358 if last_seq is None:
2359 no_fragment_score += 2
2360 last_segment_url = None
2361 continue
2362 else:
2363 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2364 no_fragment_score += 2
2365 if not should_continue:
2366 continue
2367
2368 if known_idx > last_seq:
2369 last_segment_url = None
2370 continue
2371
2372 last_seq += 1
2373
2374 if begin_index < 0 and known_idx < 0:
2375 # skip from the start when it's negative value
2376 known_idx = last_seq + begin_index
2377 if lack_early_segments:
2378 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2379 try:
2380 for idx in range(known_idx, last_seq):
2381 # do not update sequence here or you'll get skipped some part of it
2382 should_continue, _ = _extract_sequence_from_mpd(False, False)
2383 if not should_continue:
2384 known_idx = idx - 1
2385 raise ExtractorError('breaking out of outer loop')
2386 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2387 yield {
2388 'url': last_segment_url,
2389 }
2390 if known_idx == last_seq:
2391 no_fragment_score += 5
2392 else:
2393 no_fragment_score = 0
2394 known_idx = last_seq
2395 except ExtractorError:
2396 continue
2397
2398 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2399
2400 def _extract_player_url(self, *ytcfgs, webpage=None):
2401 player_url = traverse_obj(
2402 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2403 get_all=False, expected_type=compat_str)
2404 if not player_url:
2405 return
2406 return urljoin('https://www.youtube.com', player_url)
2407
2408 def _download_player_url(self, video_id, fatal=False):
2409 res = self._download_webpage(
2410 'https://www.youtube.com/iframe_api',
2411 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2412 if res:
2413 player_version = self._search_regex(
2414 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2415 if player_version:
2416 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2417
2418 def _signature_cache_id(self, example_sig):
2419 """ Return a string representation of a signature """
2420 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2421
2422 @classmethod
2423 def _extract_player_info(cls, player_url):
2424 for player_re in cls._PLAYER_INFO_RE:
2425 id_m = re.search(player_re, player_url)
2426 if id_m:
2427 break
2428 else:
2429 raise ExtractorError('Cannot identify player %r' % player_url)
2430 return id_m.group('id')
2431
2432 def _load_player(self, video_id, player_url, fatal=True):
2433 player_id = self._extract_player_info(player_url)
2434 if player_id not in self._code_cache:
2435 code = self._download_webpage(
2436 player_url, video_id, fatal=fatal,
2437 note='Downloading player ' + player_id,
2438 errnote='Download of %s failed' % player_url)
2439 if code:
2440 self._code_cache[player_id] = code
2441 return self._code_cache.get(player_id)
2442
2443 def _extract_signature_function(self, video_id, player_url, example_sig):
2444 player_id = self._extract_player_info(player_url)
2445
2446 # Read from filesystem cache
2447 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2448 assert os.path.basename(func_id) == func_id
2449
2450 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2451 if cache_spec is not None:
2452 return lambda s: ''.join(s[i] for i in cache_spec)
2453
2454 code = self._load_player(video_id, player_url)
2455 if code:
2456 res = self._parse_sig_js(code)
2457
2458 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2459 cache_res = res(test_string)
2460 cache_spec = [ord(c) for c in cache_res]
2461
2462 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2463 return res
2464
2465 def _print_sig_code(self, func, example_sig):
2466 if not self.get_param('youtube_print_sig_code'):
2467 return
2468
2469 def gen_sig_code(idxs):
2470 def _genslice(start, end, step):
2471 starts = '' if start == 0 else str(start)
2472 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2473 steps = '' if step == 1 else (':%d' % step)
2474 return f's[{starts}{ends}{steps}]'
2475
2476 step = None
2477 # Quelch pyflakes warnings - start will be set when step is set
2478 start = '(Never used)'
2479 for i, prev in zip(idxs[1:], idxs[:-1]):
2480 if step is not None:
2481 if i - prev == step:
2482 continue
2483 yield _genslice(start, prev, step)
2484 step = None
2485 continue
2486 if i - prev in [-1, 1]:
2487 step = i - prev
2488 start = prev
2489 continue
2490 else:
2491 yield 's[%d]' % prev
2492 if step is None:
2493 yield 's[%d]' % i
2494 else:
2495 yield _genslice(start, i, step)
2496
2497 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2498 cache_res = func(test_string)
2499 cache_spec = [ord(c) for c in cache_res]
2500 expr_code = ' + '.join(gen_sig_code(cache_spec))
2501 signature_id_tuple = '(%s)' % (
2502 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2503 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2504 ' return %s\n') % (signature_id_tuple, expr_code)
2505 self.to_screen('Extracted signature function:\n' + code)
2506
2507 def _parse_sig_js(self, jscode):
2508 funcname = self._search_regex(
2509 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2510 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2511 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2512 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2513 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}\(a,\d+\)',
2514 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\(\s*""\s*\)',
2515 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2516 # Obsolete patterns
2517 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2518 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2519 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2520 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2521 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2522 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2523 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2524 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2525 jscode, 'Initial JS player signature function name', group='sig')
2526
2527 jsi = JSInterpreter(jscode)
2528 initial_function = jsi.extract_function(funcname)
2529 return lambda s: initial_function([s])
2530
2531 def _decrypt_signature(self, s, video_id, player_url):
2532 """Turn the encrypted s field into a working signature"""
2533 try:
2534 player_id = (player_url, self._signature_cache_id(s))
2535 if player_id not in self._player_cache:
2536 func = self._extract_signature_function(video_id, player_url, s)
2537 self._player_cache[player_id] = func
2538 func = self._player_cache[player_id]
2539 self._print_sig_code(func, s)
2540 return func(s)
2541 except Exception as e:
2542 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2543
2544 def _decrypt_nsig(self, s, video_id, player_url):
2545 """Turn the encrypted n field into a working signature"""
2546 if player_url is None:
2547 raise ExtractorError('Cannot decrypt nsig without player_url')
2548 player_url = urljoin('https://www.youtube.com', player_url)
2549
2550 sig_id = ('nsig_value', s)
2551 if sig_id in self._player_cache:
2552 return self._player_cache[sig_id]
2553
2554 try:
2555 player_id = ('nsig', player_url)
2556 if player_id not in self._player_cache:
2557 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2558 func = self._player_cache[player_id]
2559 self._player_cache[sig_id] = func(s)
2560 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2561 return self._player_cache[sig_id]
2562 except Exception as e:
2563 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2564
2565 def _extract_n_function_name(self, jscode):
2566 nfunc, idx = self._search_regex(
2567 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2568 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2569 if not idx:
2570 return nfunc
2571 return json.loads(js_to_json(self._search_regex(
2572 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2573 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2574
2575 def _extract_n_function(self, video_id, player_url):
2576 player_id = self._extract_player_info(player_url)
2577 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2578
2579 if func_code:
2580 jsi = JSInterpreter(func_code)
2581 else:
2582 jscode = self._load_player(video_id, player_url)
2583 funcname = self._extract_n_function_name(jscode)
2584 jsi = JSInterpreter(jscode)
2585 func_code = jsi.extract_function_code(funcname)
2586 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2587
2588 if self.get_param('youtube_print_sig_code'):
2589 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2590
2591 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2592
2593 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2594 """
2595 Extract signatureTimestamp (sts)
2596 Required to tell API what sig/player version is in use.
2597 """
2598 sts = None
2599 if isinstance(ytcfg, dict):
2600 sts = int_or_none(ytcfg.get('STS'))
2601
2602 if not sts:
2603 # Attempt to extract from player
2604 if player_url is None:
2605 error_msg = 'Cannot extract signature timestamp without player_url.'
2606 if fatal:
2607 raise ExtractorError(error_msg)
2608 self.report_warning(error_msg)
2609 return
2610 code = self._load_player(video_id, player_url, fatal=fatal)
2611 if code:
2612 sts = int_or_none(self._search_regex(
2613 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2614 'JS player signature timestamp', group='sts', fatal=fatal))
2615 return sts
2616
2617 def _mark_watched(self, video_id, player_responses):
2618 playback_url = get_first(
2619 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2620 expected_type=url_or_none)
2621 if not playback_url:
2622 self.report_warning('Unable to mark watched')
2623 return
2624 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2625 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2626
2627 # cpn generation algorithm is reverse engineered from base.js.
2628 # In fact it works even with dummy cpn.
2629 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2630 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2631
2632 qs.update({
2633 'ver': ['2'],
2634 'cpn': [cpn],
2635 })
2636 playback_url = compat_urlparse.urlunparse(
2637 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2638
2639 self._download_webpage(
2640 playback_url, video_id, 'Marking watched',
2641 'Unable to mark watched', fatal=False)
2642
2643 @staticmethod
2644 def _extract_urls(webpage):
2645 # Embedded YouTube player
2646 entries = [
2647 unescapeHTML(mobj.group('url'))
2648 for mobj in re.finditer(r'''(?x)
2649 (?:
2650 <iframe[^>]+?src=|
2651 data-video-url=|
2652 <embed[^>]+?src=|
2653 embedSWF\(?:\s*|
2654 <object[^>]+data=|
2655 new\s+SWFObject\(
2656 )
2657 (["\'])
2658 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2659 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2660 \1''', webpage)]
2661
2662 # lazyYT YouTube embed
2663 entries.extend(list(map(
2664 unescapeHTML,
2665 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2666
2667 # Wordpress "YouTube Video Importer" plugin
2668 matches = re.findall(r'''(?x)<div[^>]+
2669 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2670 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2671 entries.extend(m[-1] for m in matches)
2672
2673 return entries
2674
2675 @staticmethod
2676 def _extract_url(webpage):
2677 urls = YoutubeIE._extract_urls(webpage)
2678 return urls[0] if urls else None
2679
2680 @classmethod
2681 def extract_id(cls, url):
2682 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2683 if mobj is None:
2684 raise ExtractorError('Invalid URL: %s' % url)
2685 return mobj.group('id')
2686
2687 def _extract_chapters_from_json(self, data, duration):
2688 chapter_list = traverse_obj(
2689 data, (
2690 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2691 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2692 ), expected_type=list)
2693
2694 return self._extract_chapters(
2695 chapter_list,
2696 chapter_time=lambda chapter: float_or_none(
2697 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2698 chapter_title=lambda chapter: traverse_obj(
2699 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2700 duration=duration)
2701
2702 def _extract_chapters_from_engagement_panel(self, data, duration):
2703 content_list = traverse_obj(
2704 data,
2705 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2706 expected_type=list, default=[])
2707 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2708 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2709
2710 return next((
2711 filter(None, (
2712 self._extract_chapters(
2713 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2714 chapter_time, chapter_title, duration)
2715 for contents in content_list
2716 ))), [])
2717
2718 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2719 chapters = []
2720 last_chapter = {'start_time': 0}
2721 for idx, chapter in enumerate(chapter_list or []):
2722 title = chapter_title(chapter)
2723 start_time = chapter_time(chapter)
2724 if start_time is None:
2725 continue
2726 last_chapter['end_time'] = start_time
2727 if start_time < last_chapter['start_time']:
2728 if idx == 1:
2729 chapters.pop()
2730 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2731 else:
2732 self.report_warning(f'Invalid start time for chapter "{title}"')
2733 continue
2734 last_chapter = {'start_time': start_time, 'title': title}
2735 chapters.append(last_chapter)
2736 last_chapter['end_time'] = duration
2737 return chapters
2738
2739 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2740 return self._parse_json(self._search_regex(
2741 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
2742 regex), webpage, name, default='{}'), video_id, fatal=False)
2743
2744 def _extract_comment(self, comment_renderer, parent=None):
2745 comment_id = comment_renderer.get('commentId')
2746 if not comment_id:
2747 return
2748
2749 text = self._get_text(comment_renderer, 'contentText')
2750
2751 # note: timestamp is an estimate calculated from the current time and time_text
2752 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2753 author = self._get_text(comment_renderer, 'authorText')
2754 author_id = try_get(comment_renderer,
2755 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2756
2757 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2758 lambda x: x['likeCount']), compat_str)) or 0
2759 author_thumbnail = try_get(comment_renderer,
2760 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2761
2762 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2763 is_favorited = 'creatorHeart' in (try_get(
2764 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2765 return {
2766 'id': comment_id,
2767 'text': text,
2768 'timestamp': timestamp,
2769 'time_text': time_text,
2770 'like_count': votes,
2771 'is_favorited': is_favorited,
2772 'author': author,
2773 'author_id': author_id,
2774 'author_thumbnail': author_thumbnail,
2775 'author_is_uploader': author_is_uploader,
2776 'parent': parent or 'root'
2777 }
2778
2779 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2780
2781 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2782
2783 def extract_header(contents):
2784 _continuation = None
2785 for content in contents:
2786 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2787 expected_comment_count = self._get_count(
2788 comments_header_renderer, 'countText', 'commentsCount')
2789
2790 if expected_comment_count:
2791 tracker['est_total'] = expected_comment_count
2792 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2793 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2794
2795 sort_menu_item = try_get(
2796 comments_header_renderer,
2797 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2798 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2799
2800 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2801 if not _continuation:
2802 continue
2803
2804 sort_text = str_or_none(sort_menu_item.get('title'))
2805 if not sort_text:
2806 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2807 self.to_screen('Sorting comments by %s' % sort_text.lower())
2808 break
2809 return _continuation
2810
2811 def extract_thread(contents):
2812 if not parent:
2813 tracker['current_page_thread'] = 0
2814 for content in contents:
2815 if not parent and tracker['total_parent_comments'] >= max_parents:
2816 yield
2817 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2818 comment_renderer = get_first(
2819 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2820 expected_type=dict, default={})
2821
2822 comment = self._extract_comment(comment_renderer, parent)
2823 if not comment:
2824 continue
2825
2826 tracker['running_total'] += 1
2827 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2828 yield comment
2829
2830 # Attempt to get the replies
2831 comment_replies_renderer = try_get(
2832 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2833
2834 if comment_replies_renderer:
2835 tracker['current_page_thread'] += 1
2836 comment_entries_iter = self._comment_entries(
2837 comment_replies_renderer, ytcfg, video_id,
2838 parent=comment.get('id'), tracker=tracker)
2839 yield from itertools.islice(comment_entries_iter, min(
2840 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2841
2842 # Keeps track of counts across recursive calls
2843 if not tracker:
2844 tracker = dict(
2845 running_total=0,
2846 est_total=0,
2847 current_page_thread=0,
2848 total_parent_comments=0,
2849 total_reply_comments=0)
2850
2851 # TODO: Deprecated
2852 # YouTube comments have a max depth of 2
2853 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2854 if max_depth:
2855 self._downloader.deprecation_warning(
2856 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2857 if max_depth == 1 and parent:
2858 return
2859
2860 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2861 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2862
2863 continuation = self._extract_continuation(root_continuation_data)
2864
2865 response = None
2866 is_forced_continuation = False
2867 is_first_continuation = parent is None
2868 if is_first_continuation and not continuation:
2869 # Sometimes you can get comments by generating the continuation yourself,
2870 # even if YouTube initially reports them being disabled - e.g. stories comments.
2871 # Note: if the comment section is actually disabled, YouTube may return a response with
2872 # required check_get_keys missing. So we will disable that check initially in this case.
2873 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2874 is_forced_continuation = True
2875
2876 for page_num in itertools.count(0):
2877 if not continuation:
2878 break
2879 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2880 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2881 if page_num == 0:
2882 if is_first_continuation:
2883 note_prefix = 'Downloading comment section API JSON'
2884 else:
2885 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2886 tracker['current_page_thread'], comment_prog_str)
2887 else:
2888 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2889 ' ' if parent else '', ' replies' if parent else '',
2890 page_num, comment_prog_str)
2891
2892 response = self._extract_response(
2893 item_id=None, query=continuation,
2894 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2895 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2896 is_forced_continuation = False
2897 continuation_contents = traverse_obj(
2898 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2899
2900 continuation = None
2901 for continuation_section in continuation_contents:
2902 continuation_items = traverse_obj(
2903 continuation_section,
2904 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2905 get_all=False, expected_type=list) or []
2906 if is_first_continuation:
2907 continuation = extract_header(continuation_items)
2908 is_first_continuation = False
2909 if continuation:
2910 break
2911 continue
2912
2913 for entry in extract_thread(continuation_items):
2914 if not entry:
2915 return
2916 yield entry
2917 continuation = self._extract_continuation({'contents': continuation_items})
2918 if continuation:
2919 break
2920
2921 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2922 if message and not parent and tracker['running_total'] == 0:
2923 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2924
2925 @staticmethod
2926 def _generate_comment_continuation(video_id):
2927 """
2928 Generates initial comment section continuation token from given video id
2929 """
2930 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2931 return base64.b64encode(token.encode()).decode()
2932
2933 def _get_comments(self, ytcfg, video_id, contents, webpage):
2934 """Entry for comment extraction"""
2935 def _real_comment_extract(contents):
2936 renderer = next((
2937 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2938 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2939 yield from self._comment_entries(renderer, ytcfg, video_id)
2940
2941 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2942 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2943
2944 @staticmethod
2945 def _get_checkok_params():
2946 return {'contentCheckOk': True, 'racyCheckOk': True}
2947
2948 @classmethod
2949 def _generate_player_context(cls, sts=None):
2950 context = {
2951 'html5Preference': 'HTML5_PREF_WANTS',
2952 }
2953 if sts is not None:
2954 context['signatureTimestamp'] = sts
2955 return {
2956 'playbackContext': {
2957 'contentPlaybackContext': context
2958 },
2959 **cls._get_checkok_params()
2960 }
2961
2962 @staticmethod
2963 def _is_agegated(player_response):
2964 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2965 return True
2966
2967 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2968 AGE_GATE_REASONS = (
2969 'confirm your age', 'age-restricted', 'inappropriate', # reason
2970 'age_verification_required', 'age_check_required', # status
2971 )
2972 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2973
2974 @staticmethod
2975 def _is_unplayable(player_response):
2976 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2977
2978 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2979
2980 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2981 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2982 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2983 headers = self.generate_api_headers(
2984 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
2985
2986 yt_query = {
2987 'videoId': video_id,
2988 'params': '8AEB' # enable stories
2989 }
2990 yt_query.update(self._generate_player_context(sts))
2991 return self._extract_response(
2992 item_id=video_id, ep='player', query=yt_query,
2993 ytcfg=player_ytcfg, headers=headers, fatal=True,
2994 default_client=client,
2995 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
2996 ) or None
2997
2998 def _get_requested_clients(self, url, smuggled_data):
2999 requested_clients = []
3000 default = ['android', 'web']
3001 allowed_clients = sorted(
3002 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3003 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3004 for client in self._configuration_arg('player_client'):
3005 if client in allowed_clients:
3006 requested_clients.append(client)
3007 elif client == 'default':
3008 requested_clients.extend(default)
3009 elif client == 'all':
3010 requested_clients.extend(allowed_clients)
3011 else:
3012 self.report_warning(f'Skipping unsupported client {client}')
3013 if not requested_clients:
3014 requested_clients = default
3015
3016 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3017 requested_clients.extend(
3018 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3019
3020 return orderedSet(requested_clients)
3021
3022 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3023 initial_pr = None
3024 if webpage:
3025 initial_pr = self._extract_yt_initial_variable(
3026 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
3027 video_id, 'initial player response')
3028
3029 all_clients = set(clients)
3030 clients = clients[::-1]
3031 prs = []
3032
3033 def append_client(*client_names):
3034 """ Append the first client name that exists but not already used """
3035 for client_name in client_names:
3036 actual_client = _split_innertube_client(client_name)[0]
3037 if actual_client in INNERTUBE_CLIENTS:
3038 if actual_client not in all_clients:
3039 clients.append(client_name)
3040 all_clients.add(actual_client)
3041 return
3042
3043 # Android player_response does not have microFormats which are needed for
3044 # extraction of some data. So we return the initial_pr with formats
3045 # stripped out even if not requested by the user
3046 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3047 if initial_pr:
3048 pr = dict(initial_pr)
3049 pr['streamingData'] = None
3050 prs.append(pr)
3051
3052 last_error = None
3053 tried_iframe_fallback = False
3054 player_url = None
3055 while clients:
3056 client, base_client, variant = _split_innertube_client(clients.pop())
3057 player_ytcfg = master_ytcfg if client == 'web' else {}
3058 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3059 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3060
3061 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3062 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3063 if 'js' in self._configuration_arg('player_skip'):
3064 require_js_player = False
3065 player_url = None
3066
3067 if not player_url and not tried_iframe_fallback and require_js_player:
3068 player_url = self._download_player_url(video_id)
3069 tried_iframe_fallback = True
3070
3071 try:
3072 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3073 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3074 except ExtractorError as e:
3075 if last_error:
3076 self.report_warning(last_error)
3077 last_error = e
3078 continue
3079
3080 if pr:
3081 prs.append(pr)
3082
3083 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3084 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3085 append_client(f'{base_client}_creator')
3086 elif self._is_agegated(pr):
3087 if variant == 'tv_embedded':
3088 append_client(f'{base_client}_embedded')
3089 elif not variant:
3090 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3091
3092 if last_error:
3093 if not len(prs):
3094 raise last_error
3095 self.report_warning(last_error)
3096 return prs, player_url
3097
3098 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3099 itags, stream_ids = {}, []
3100 itag_qualities, res_qualities = {}, {}
3101 q = qualities([
3102 # Normally tiny is the smallest video-only formats. But
3103 # audio-only formats with unknown quality may get tagged as tiny
3104 'tiny',
3105 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3106 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3107 ])
3108 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3109
3110 for fmt in streaming_formats:
3111 if fmt.get('targetDurationSec'):
3112 continue
3113
3114 itag = str_or_none(fmt.get('itag'))
3115 audio_track = fmt.get('audioTrack') or {}
3116 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3117 if stream_id in stream_ids:
3118 continue
3119
3120 quality = fmt.get('quality')
3121 height = int_or_none(fmt.get('height'))
3122 if quality == 'tiny' or not quality:
3123 quality = fmt.get('audioQuality', '').lower() or quality
3124 # The 3gp format (17) in android client has a quality of "small",
3125 # but is actually worse than other formats
3126 if itag == '17':
3127 quality = 'tiny'
3128 if quality:
3129 if itag:
3130 itag_qualities[itag] = quality
3131 if height:
3132 res_qualities[height] = quality
3133 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3134 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3135 # number of fragment that would subsequently requested with (`&sq=N`)
3136 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3137 continue
3138
3139 fmt_url = fmt.get('url')
3140 if not fmt_url:
3141 sc = compat_parse_qs(fmt.get('signatureCipher'))
3142 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3143 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3144 if not all((sc, fmt_url, player_url, encrypted_sig)):
3145 continue
3146 try:
3147 fmt_url += '&%s=%s' % (
3148 traverse_obj(sc, ('sp', -1)) or 'signature',
3149 self._decrypt_signature(encrypted_sig, video_id, player_url)
3150 )
3151 except ExtractorError as e:
3152 self.report_warning('Signature extraction failed: Some formats may be missing', only_once=True)
3153 self.write_debug(e, only_once=True)
3154 continue
3155
3156 query = parse_qs(fmt_url)
3157 throttled = False
3158 if query.get('n'):
3159 try:
3160 fmt_url = update_url_query(fmt_url, {
3161 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3162 except ExtractorError as e:
3163 self.report_warning(
3164 'nsig extraction failed: You may experience throttling for some formats\n'
3165 f'n = {query["n"][0]} ; player = {player_url}', only_once=True)
3166 self.write_debug(e, only_once=True)
3167 throttled = True
3168
3169 if itag:
3170 itags[itag] = 'https'
3171 stream_ids.append(stream_id)
3172
3173 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3174 language_preference = (
3175 10 if audio_track.get('audioIsDefault') and 10
3176 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3177 else -1)
3178 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3179 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3180 # Make sure to avoid false positives with small duration differences.
3181 # Eg: __2ABJjxzNo, ySuUZEjARPY
3182 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3183 if is_damaged:
3184 self.report_warning(
3185 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3186 dct = {
3187 'asr': int_or_none(fmt.get('audioSampleRate')),
3188 'filesize': int_or_none(fmt.get('contentLength')),
3189 'format_id': itag,
3190 'format_note': join_nonempty(
3191 '%s%s' % (audio_track.get('displayName') or '',
3192 ' (default)' if language_preference > 0 else ''),
3193 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3194 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3195 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3196 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3197 'fps': int_or_none(fmt.get('fps')) or None,
3198 'height': height,
3199 'quality': q(quality),
3200 'has_drm': bool(fmt.get('drmFamilies')),
3201 'tbr': tbr,
3202 'url': fmt_url,
3203 'width': int_or_none(fmt.get('width')),
3204 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3205 'desc' if language_preference < -1 else ''),
3206 'language_preference': language_preference,
3207 # Strictly de-prioritize damaged and 3gp formats
3208 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3209 }
3210 mime_mobj = re.match(
3211 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3212 if mime_mobj:
3213 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3214 dct.update(parse_codecs(mime_mobj.group(2)))
3215 no_audio = dct.get('acodec') == 'none'
3216 no_video = dct.get('vcodec') == 'none'
3217 if no_audio:
3218 dct['vbr'] = tbr
3219 if no_video:
3220 dct['abr'] = tbr
3221 if no_audio or no_video:
3222 dct['downloader_options'] = {
3223 # Youtube throttles chunks >~10M
3224 'http_chunk_size': 10485760,
3225 }
3226 if dct.get('ext'):
3227 dct['container'] = dct['ext'] + '_dash'
3228 yield dct
3229
3230 live_from_start = is_live and self.get_param('live_from_start')
3231 skip_manifests = self._configuration_arg('skip')
3232 if not self.get_param('youtube_include_hls_manifest', True):
3233 skip_manifests.append('hls')
3234 if not self.get_param('youtube_include_dash_manifest', True):
3235 skip_manifests.append('dash')
3236 get_dash = 'dash' not in skip_manifests and (
3237 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3238 get_hls = not live_from_start and 'hls' not in skip_manifests
3239
3240 def process_manifest_format(f, proto, itag):
3241 if itag in itags:
3242 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3243 return False
3244 itag = f'{itag}-{proto}'
3245 if itag:
3246 f['format_id'] = itag
3247 itags[itag] = proto
3248
3249 f['quality'] = next((
3250 q(qdict[val])
3251 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3252 if val in qdict), -1)
3253 return True
3254
3255 for sd in streaming_data:
3256 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3257 if hls_manifest_url:
3258 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3259 if process_manifest_format(f, 'hls', self._search_regex(
3260 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3261 yield f
3262
3263 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3264 if dash_manifest_url:
3265 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3266 if process_manifest_format(f, 'dash', f['format_id']):
3267 f['filesize'] = int_or_none(self._search_regex(
3268 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3269 if live_from_start:
3270 f['is_from_start'] = True
3271
3272 yield f
3273
3274 def _extract_storyboard(self, player_responses, duration):
3275 spec = get_first(
3276 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3277 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3278 if not base_url:
3279 return
3280 L = len(spec) - 1
3281 for i, args in enumerate(spec):
3282 args = args.split('#')
3283 counts = list(map(int_or_none, args[:5]))
3284 if len(args) != 8 or not all(counts):
3285 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3286 continue
3287 width, height, frame_count, cols, rows = counts
3288 N, sigh = args[6:]
3289
3290 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3291 fragment_count = frame_count / (cols * rows)
3292 fragment_duration = duration / fragment_count
3293 yield {
3294 'format_id': f'sb{i}',
3295 'format_note': 'storyboard',
3296 'ext': 'mhtml',
3297 'protocol': 'mhtml',
3298 'acodec': 'none',
3299 'vcodec': 'none',
3300 'url': url,
3301 'width': width,
3302 'height': height,
3303 'fragments': [{
3304 'url': url.replace('$M', str(j)),
3305 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3306 } for j in range(math.ceil(fragment_count))],
3307 }
3308
3309 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3310 webpage = None
3311 if 'webpage' not in self._configuration_arg('player_skip'):
3312 webpage = self._download_webpage(
3313 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3314
3315 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3316
3317 player_responses, player_url = self._extract_player_responses(
3318 self._get_requested_clients(url, smuggled_data),
3319 video_id, webpage, master_ytcfg)
3320
3321 return webpage, master_ytcfg, player_responses, player_url
3322
3323 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3324 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3325 is_live = get_first(video_details, 'isLive')
3326 if is_live is None:
3327 is_live = get_first(live_broadcast_details, 'isLiveNow')
3328
3329 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3330 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3331
3332 return live_broadcast_details, is_live, streaming_data, formats
3333
3334 def _real_extract(self, url):
3335 url, smuggled_data = unsmuggle_url(url, {})
3336 video_id = self._match_id(url)
3337
3338 base_url = self.http_scheme() + '//www.youtube.com/'
3339 webpage_url = base_url + 'watch?v=' + video_id
3340
3341 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3342
3343 playability_statuses = traverse_obj(
3344 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3345
3346 trailer_video_id = get_first(
3347 playability_statuses,
3348 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3349 expected_type=str)
3350 if trailer_video_id:
3351 return self.url_result(
3352 trailer_video_id, self.ie_key(), trailer_video_id)
3353
3354 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3355 if webpage else (lambda x: None))
3356
3357 video_details = traverse_obj(
3358 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3359 microformats = traverse_obj(
3360 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3361 expected_type=dict, default=[])
3362 video_title = (
3363 get_first(video_details, 'title')
3364 or self._get_text(microformats, (..., 'title'))
3365 or search_meta(['og:title', 'twitter:title', 'title']))
3366 video_description = get_first(video_details, 'shortDescription')
3367
3368 multifeed_metadata_list = get_first(
3369 player_responses,
3370 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3371 expected_type=str)
3372 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3373 if self.get_param('noplaylist'):
3374 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3375 else:
3376 entries = []
3377 feed_ids = []
3378 for feed in multifeed_metadata_list.split(','):
3379 # Unquote should take place before split on comma (,) since textual
3380 # fields may contain comma as well (see
3381 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3382 feed_data = compat_parse_qs(
3383 compat_urllib_parse_unquote_plus(feed))
3384
3385 def feed_entry(name):
3386 return try_get(
3387 feed_data, lambda x: x[name][0], compat_str)
3388
3389 feed_id = feed_entry('id')
3390 if not feed_id:
3391 continue
3392 feed_title = feed_entry('title')
3393 title = video_title
3394 if feed_title:
3395 title += ' (%s)' % feed_title
3396 entries.append({
3397 '_type': 'url_transparent',
3398 'ie_key': 'Youtube',
3399 'url': smuggle_url(
3400 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3401 {'force_singlefeed': True}),
3402 'title': title,
3403 })
3404 feed_ids.append(feed_id)
3405 self.to_screen(
3406 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3407 % (', '.join(feed_ids), video_id))
3408 return self.playlist_result(
3409 entries, video_id, video_title, video_description)
3410
3411 duration = int_or_none(
3412 get_first(video_details, 'lengthSeconds')
3413 or get_first(microformats, 'lengthSeconds')
3414 or parse_duration(search_meta('duration'))) or None
3415
3416 if get_first(video_details, 'isPostLiveDvr'):
3417 self.write_debug('Video is in Post-Live Manifestless mode')
3418 if duration or 0 > 4 * 3600:
3419 self.report_warning(
3420 'The livestream has not finished processing. Only 4 hours of the video can be currently downloaded. '
3421 'This is a known issue and patches are welcome')
3422
3423 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3424 video_id, microformats, video_details, player_responses, player_url, duration)
3425
3426 if not formats:
3427 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3428 self.report_drm(video_id)
3429 pemr = get_first(
3430 playability_statuses,
3431 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3432 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3433 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3434 if subreason:
3435 if subreason == 'The uploader has not made this video available in your country.':
3436 countries = get_first(microformats, 'availableCountries')
3437 if not countries:
3438 regions_allowed = search_meta('regionsAllowed')
3439 countries = regions_allowed.split(',') if regions_allowed else None
3440 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3441 reason += f'. {subreason}'
3442 if reason:
3443 self.raise_no_formats(reason, expected=True)
3444
3445 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3446 if not keywords and webpage:
3447 keywords = [
3448 unescapeHTML(m.group('content'))
3449 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3450 for keyword in keywords:
3451 if keyword.startswith('yt:stretch='):
3452 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3453 if mobj:
3454 # NB: float is intentional for forcing float division
3455 w, h = (float(v) for v in mobj.groups())
3456 if w > 0 and h > 0:
3457 ratio = w / h
3458 for f in formats:
3459 if f.get('vcodec') != 'none':
3460 f['stretched_ratio'] = ratio
3461 break
3462 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3463 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3464 if thumbnail_url:
3465 thumbnails.append({
3466 'url': thumbnail_url,
3467 })
3468 original_thumbnails = thumbnails.copy()
3469
3470 # The best resolution thumbnails sometimes does not appear in the webpage
3471 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3472 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3473 thumbnail_names = [
3474 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3475 # in resolution, these are not the custom thumbnail. So de-prioritize them
3476 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3477 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3478 ]
3479 n_thumbnail_names = len(thumbnail_names)
3480 thumbnails.extend({
3481 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3482 video_id=video_id, name=name, ext=ext,
3483 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3484 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3485 for thumb in thumbnails:
3486 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3487 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3488 self._remove_duplicate_formats(thumbnails)
3489 self._downloader._sort_thumbnails(original_thumbnails)
3490
3491 category = get_first(microformats, 'category') or search_meta('genre')
3492 channel_id = str_or_none(
3493 get_first(video_details, 'channelId')
3494 or get_first(microformats, 'externalChannelId')
3495 or search_meta('channelId'))
3496 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3497
3498 live_content = get_first(video_details, 'isLiveContent')
3499 is_upcoming = get_first(video_details, 'isUpcoming')
3500 if is_live is None:
3501 if is_upcoming or live_content is False:
3502 is_live = False
3503 if is_upcoming is None and (live_content or is_live):
3504 is_upcoming = False
3505 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3506 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3507 if not duration and live_end_time and live_start_time:
3508 duration = live_end_time - live_start_time
3509
3510 if is_live and self.get_param('live_from_start'):
3511 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3512
3513 formats.extend(self._extract_storyboard(player_responses, duration))
3514
3515 # Source is given priority since formats that throttle are given lower source_preference
3516 # When throttling issue is fully fixed, remove this
3517 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3518
3519 info = {
3520 'id': video_id,
3521 'title': video_title,
3522 'formats': formats,
3523 'thumbnails': thumbnails,
3524 # The best thumbnail that we are sure exists. Prevents unnecessary
3525 # URL checking if user don't care about getting the best possible thumbnail
3526 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3527 'description': video_description,
3528 'uploader': get_first(video_details, 'author'),
3529 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3530 'uploader_url': owner_profile_url,
3531 'channel_id': channel_id,
3532 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3533 'duration': duration,
3534 'view_count': int_or_none(
3535 get_first((video_details, microformats), (..., 'viewCount'))
3536 or search_meta('interactionCount')),
3537 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3538 'age_limit': 18 if (
3539 get_first(microformats, 'isFamilySafe') is False
3540 or search_meta('isFamilyFriendly') == 'false'
3541 or search_meta('og:restrictions:age') == '18+') else 0,
3542 'webpage_url': webpage_url,
3543 'categories': [category] if category else None,
3544 'tags': keywords,
3545 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3546 'is_live': is_live,
3547 'was_live': (False if is_live or is_upcoming or live_content is False
3548 else None if is_live is None or is_upcoming is None
3549 else live_content),
3550 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3551 'release_timestamp': live_start_time,
3552 }
3553
3554 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3555 if pctr:
3556 def get_lang_code(track):
3557 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3558 or track.get('languageCode'))
3559
3560 # Converted into dicts to remove duplicates
3561 captions = {
3562 get_lang_code(sub): sub
3563 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3564 translation_languages = {
3565 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3566 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3567
3568 def process_language(container, base_url, lang_code, sub_name, query):
3569 lang_subs = container.setdefault(lang_code, [])
3570 for fmt in self._SUBTITLE_FORMATS:
3571 query.update({
3572 'fmt': fmt,
3573 })
3574 lang_subs.append({
3575 'ext': fmt,
3576 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3577 'name': sub_name,
3578 })
3579
3580 subtitles, automatic_captions = {}, {}
3581 for lang_code, caption_track in captions.items():
3582 base_url = caption_track.get('baseUrl')
3583 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3584 if not base_url:
3585 continue
3586 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3587 if caption_track.get('kind') != 'asr':
3588 if not lang_code:
3589 continue
3590 process_language(
3591 subtitles, base_url, lang_code, lang_name, {})
3592 if not caption_track.get('isTranslatable'):
3593 continue
3594 for trans_code, trans_name in translation_languages.items():
3595 if not trans_code:
3596 continue
3597 orig_trans_code = trans_code
3598 if caption_track.get('kind') != 'asr':
3599 if 'translated_subs' in self._configuration_arg('skip'):
3600 continue
3601 trans_code += f'-{lang_code}'
3602 trans_name += format_field(lang_name, template=' from %s')
3603 # Add an "-orig" label to the original language so that it can be distinguished.
3604 # The subs are returned without "-orig" as well for compatibility
3605 if lang_code == f'a-{orig_trans_code}':
3606 process_language(
3607 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3608 # Setting tlang=lang returns damaged subtitles.
3609 process_language(automatic_captions, base_url, trans_code, trans_name,
3610 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3611 info['automatic_captions'] = automatic_captions
3612 info['subtitles'] = subtitles
3613
3614 parsed_url = compat_urllib_parse_urlparse(url)
3615 for component in [parsed_url.fragment, parsed_url.query]:
3616 query = compat_parse_qs(component)
3617 for k, v in query.items():
3618 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3619 d_k += '_time'
3620 if d_k not in info and k in s_ks:
3621 info[d_k] = parse_duration(query[k][0])
3622
3623 # Youtube Music Auto-generated description
3624 if video_description:
3625 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)
3626 if mobj:
3627 release_year = mobj.group('release_year')
3628 release_date = mobj.group('release_date')
3629 if release_date:
3630 release_date = release_date.replace('-', '')
3631 if not release_year:
3632 release_year = release_date[:4]
3633 info.update({
3634 'album': mobj.group('album'.strip()),
3635 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3636 'track': mobj.group('track').strip(),
3637 'release_date': release_date,
3638 'release_year': int_or_none(release_year),
3639 })
3640
3641 initial_data = None
3642 if webpage:
3643 initial_data = self._extract_yt_initial_variable(
3644 webpage, self._YT_INITIAL_DATA_RE, video_id,
3645 'yt initial data')
3646 if not initial_data:
3647 query = {'videoId': video_id}
3648 query.update(self._get_checkok_params())
3649 initial_data = self._extract_response(
3650 item_id=video_id, ep='next', fatal=False,
3651 ytcfg=master_ytcfg, query=query,
3652 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3653 note='Downloading initial data API JSON')
3654
3655 try: # This will error if there is no livechat
3656 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3657 except (KeyError, IndexError, TypeError):
3658 pass
3659 else:
3660 info.setdefault('subtitles', {})['live_chat'] = [{
3661 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3662 'video_id': video_id,
3663 'ext': 'json',
3664 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3665 }]
3666
3667 if initial_data:
3668 info['chapters'] = (
3669 self._extract_chapters_from_json(initial_data, duration)
3670 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3671 or None)
3672
3673 contents = traverse_obj(
3674 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3675 expected_type=list, default=[])
3676
3677 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3678 if vpir:
3679 stl = vpir.get('superTitleLink')
3680 if stl:
3681 stl = self._get_text(stl)
3682 if try_get(
3683 vpir,
3684 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3685 info['location'] = stl
3686 else:
3687 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3688 if mobj:
3689 info.update({
3690 'series': mobj.group(1),
3691 'season_number': int(mobj.group(2)),
3692 'episode_number': int(mobj.group(3)),
3693 })
3694 for tlb in (try_get(
3695 vpir,
3696 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3697 list) or []):
3698 tbr = tlb.get('toggleButtonRenderer') or {}
3699 for getter, regex in [(
3700 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3701 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3702 lambda x: x['accessibility'],
3703 lambda x: x['accessibilityData']['accessibilityData'],
3704 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3705 label = (try_get(tbr, getter, dict) or {}).get('label')
3706 if label:
3707 mobj = re.match(regex, label)
3708 if mobj:
3709 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3710 break
3711 sbr_tooltip = try_get(
3712 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3713 if sbr_tooltip:
3714 like_count, dislike_count = sbr_tooltip.split(' / ')
3715 info.update({
3716 'like_count': str_to_int(like_count),
3717 'dislike_count': str_to_int(dislike_count),
3718 })
3719 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3720 if vsir:
3721 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3722 info.update({
3723 'channel': self._get_text(vor, 'title'),
3724 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3725
3726 rows = try_get(
3727 vsir,
3728 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3729 list) or []
3730 multiple_songs = False
3731 for row in rows:
3732 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3733 multiple_songs = True
3734 break
3735 for row in rows:
3736 mrr = row.get('metadataRowRenderer') or {}
3737 mrr_title = mrr.get('title')
3738 if not mrr_title:
3739 continue
3740 mrr_title = self._get_text(mrr, 'title')
3741 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3742 if mrr_title == 'License':
3743 info['license'] = mrr_contents_text
3744 elif not multiple_songs:
3745 if mrr_title == 'Album':
3746 info['album'] = mrr_contents_text
3747 elif mrr_title == 'Artist':
3748 info['artist'] = mrr_contents_text
3749 elif mrr_title == 'Song':
3750 info['track'] = mrr_contents_text
3751
3752 fallbacks = {
3753 'channel': 'uploader',
3754 'channel_id': 'uploader_id',
3755 'channel_url': 'uploader_url',
3756 }
3757
3758 # The upload date for scheduled, live and past live streams / premieres in microformats
3759 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3760 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3761 upload_date = (
3762 unified_strdate(get_first(microformats, 'uploadDate'))
3763 or unified_strdate(search_meta('uploadDate')))
3764 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3765 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3766 info['upload_date'] = upload_date
3767
3768 for to, frm in fallbacks.items():
3769 if not info.get(to):
3770 info[to] = info.get(frm)
3771
3772 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3773 v = info.get(s_k)
3774 if v:
3775 info[d_k] = v
3776
3777 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3778 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3779 is_membersonly = None
3780 is_premium = None
3781 if initial_data and is_private is not None:
3782 is_membersonly = False
3783 is_premium = False
3784 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3785 badge_labels = set()
3786 for content in contents:
3787 if not isinstance(content, dict):
3788 continue
3789 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3790 for badge_label in badge_labels:
3791 if badge_label.lower() == 'members only':
3792 is_membersonly = True
3793 elif badge_label.lower() == 'premium':
3794 is_premium = True
3795 elif badge_label.lower() == 'unlisted':
3796 is_unlisted = True
3797
3798 info['availability'] = self._availability(
3799 is_private=is_private,
3800 needs_premium=is_premium,
3801 needs_subscription=is_membersonly,
3802 needs_auth=info['age_limit'] >= 18,
3803 is_unlisted=None if is_private is None else is_unlisted)
3804
3805 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3806
3807 self.mark_watched(video_id, player_responses)
3808
3809 return info
3810
3811
3812class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3813
3814 @staticmethod
3815 def passthrough_smuggled_data(func):
3816 def _smuggle(entries, smuggled_data):
3817 for entry in entries:
3818 # TODO: Convert URL to music.youtube instead.
3819 # Do we need to passthrough any other smuggled_data?
3820 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3821 yield entry
3822
3823 @functools.wraps(func)
3824 def wrapper(self, url):
3825 url, smuggled_data = unsmuggle_url(url, {})
3826 if self.is_music_url(url):
3827 smuggled_data['is_music_url'] = True
3828 info_dict = func(self, url, smuggled_data)
3829 if smuggled_data and info_dict.get('entries'):
3830 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3831 return info_dict
3832 return wrapper
3833
3834 def _extract_channel_id(self, webpage):
3835 channel_id = self._html_search_meta(
3836 'channelId', webpage, 'channel id', default=None)
3837 if channel_id:
3838 return channel_id
3839 channel_url = self._html_search_meta(
3840 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3841 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3842 'twitter:app:url:googleplay'), webpage, 'channel url')
3843 return self._search_regex(
3844 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3845 channel_url, 'channel id')
3846
3847 @staticmethod
3848 def _extract_basic_item_renderer(item):
3849 # Modified from _extract_grid_item_renderer
3850 known_basic_renderers = (
3851 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3852 )
3853 for key, renderer in item.items():
3854 if not isinstance(renderer, dict):
3855 continue
3856 elif key in known_basic_renderers:
3857 return renderer
3858 elif key.startswith('grid') and key.endswith('Renderer'):
3859 return renderer
3860
3861 def _grid_entries(self, grid_renderer):
3862 for item in grid_renderer['items']:
3863 if not isinstance(item, dict):
3864 continue
3865 renderer = self._extract_basic_item_renderer(item)
3866 if not isinstance(renderer, dict):
3867 continue
3868 title = self._get_text(renderer, 'title')
3869
3870 # playlist
3871 playlist_id = renderer.get('playlistId')
3872 if playlist_id:
3873 yield self.url_result(
3874 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3875 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3876 video_title=title)
3877 continue
3878 # video
3879 video_id = renderer.get('videoId')
3880 if video_id:
3881 yield self._extract_video(renderer)
3882 continue
3883 # channel
3884 channel_id = renderer.get('channelId')
3885 if channel_id:
3886 yield self.url_result(
3887 'https://www.youtube.com/channel/%s' % channel_id,
3888 ie=YoutubeTabIE.ie_key(), video_title=title)
3889 continue
3890 # generic endpoint URL support
3891 ep_url = urljoin('https://www.youtube.com/', try_get(
3892 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3893 compat_str))
3894 if ep_url:
3895 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3896 if ie.suitable(ep_url):
3897 yield self.url_result(
3898 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3899 break
3900
3901 def _music_reponsive_list_entry(self, renderer):
3902 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3903 if video_id:
3904 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3905 ie=YoutubeIE.ie_key(), video_id=video_id)
3906 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3907 if playlist_id:
3908 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3909 if video_id:
3910 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3911 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3912 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3913 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3914 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3915 if browse_id:
3916 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3917 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3918
3919 def _shelf_entries_from_content(self, shelf_renderer):
3920 content = shelf_renderer.get('content')
3921 if not isinstance(content, dict):
3922 return
3923 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3924 if renderer:
3925 # TODO: add support for nested playlists so each shelf is processed
3926 # as separate playlist
3927 # TODO: this includes only first N items
3928 yield from self._grid_entries(renderer)
3929 renderer = content.get('horizontalListRenderer')
3930 if renderer:
3931 # TODO
3932 pass
3933
3934 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3935 ep = try_get(
3936 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3937 compat_str)
3938 shelf_url = urljoin('https://www.youtube.com', ep)
3939 if shelf_url:
3940 # Skipping links to another channels, note that checking for
3941 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3942 # will not work
3943 if skip_channels and '/channels?' in shelf_url:
3944 return
3945 title = self._get_text(shelf_renderer, 'title')
3946 yield self.url_result(shelf_url, video_title=title)
3947 # Shelf may not contain shelf URL, fallback to extraction from content
3948 yield from self._shelf_entries_from_content(shelf_renderer)
3949
3950 def _playlist_entries(self, video_list_renderer):
3951 for content in video_list_renderer['contents']:
3952 if not isinstance(content, dict):
3953 continue
3954 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3955 if not isinstance(renderer, dict):
3956 continue
3957 video_id = renderer.get('videoId')
3958 if not video_id:
3959 continue
3960 yield self._extract_video(renderer)
3961
3962 def _rich_entries(self, rich_grid_renderer):
3963 renderer = try_get(
3964 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3965 video_id = renderer.get('videoId')
3966 if not video_id:
3967 return
3968 yield self._extract_video(renderer)
3969
3970 def _video_entry(self, video_renderer):
3971 video_id = video_renderer.get('videoId')
3972 if video_id:
3973 return self._extract_video(video_renderer)
3974
3975 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3976 url = urljoin('https://youtube.com', traverse_obj(
3977 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3978 if url:
3979 return self.url_result(
3980 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3981
3982 def _post_thread_entries(self, post_thread_renderer):
3983 post_renderer = try_get(
3984 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3985 if not post_renderer:
3986 return
3987 # video attachment
3988 video_renderer = try_get(
3989 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3990 video_id = video_renderer.get('videoId')
3991 if video_id:
3992 entry = self._extract_video(video_renderer)
3993 if entry:
3994 yield entry
3995 # playlist attachment
3996 playlist_id = try_get(
3997 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3998 if playlist_id:
3999 yield self.url_result(
4000 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4001 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4002 # inline video links
4003 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4004 for run in runs:
4005 if not isinstance(run, dict):
4006 continue
4007 ep_url = try_get(
4008 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
4009 if not ep_url:
4010 continue
4011 if not YoutubeIE.suitable(ep_url):
4012 continue
4013 ep_video_id = YoutubeIE._match_id(ep_url)
4014 if video_id == ep_video_id:
4015 continue
4016 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4017
4018 def _post_thread_continuation_entries(self, post_thread_continuation):
4019 contents = post_thread_continuation.get('contents')
4020 if not isinstance(contents, list):
4021 return
4022 for content in contents:
4023 renderer = content.get('backstagePostThreadRenderer')
4024 if not isinstance(renderer, dict):
4025 continue
4026 yield from self._post_thread_entries(renderer)
4027
4028 r''' # unused
4029 def _rich_grid_entries(self, contents):
4030 for content in contents:
4031 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4032 if video_renderer:
4033 entry = self._video_entry(video_renderer)
4034 if entry:
4035 yield entry
4036 '''
4037
4038 def _extract_entries(self, parent_renderer, continuation_list):
4039 # continuation_list is modified in-place with continuation_list = [continuation_token]
4040 continuation_list[:] = [None]
4041 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4042 for content in contents:
4043 if not isinstance(content, dict):
4044 continue
4045 is_renderer = traverse_obj(
4046 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4047 expected_type=dict)
4048 if not is_renderer:
4049 renderer = content.get('richItemRenderer')
4050 if renderer:
4051 for entry in self._rich_entries(renderer):
4052 yield entry
4053 continuation_list[0] = self._extract_continuation(parent_renderer)
4054 continue
4055 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4056 for isr_content in isr_contents:
4057 if not isinstance(isr_content, dict):
4058 continue
4059
4060 known_renderers = {
4061 'playlistVideoListRenderer': self._playlist_entries,
4062 'gridRenderer': self._grid_entries,
4063 'reelShelfRenderer': self._grid_entries,
4064 'shelfRenderer': self._shelf_entries,
4065 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4066 'backstagePostThreadRenderer': self._post_thread_entries,
4067 'videoRenderer': lambda x: [self._video_entry(x)],
4068 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4069 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4070 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4071 }
4072 for key, renderer in isr_content.items():
4073 if key not in known_renderers:
4074 continue
4075 for entry in known_renderers[key](renderer):
4076 if entry:
4077 yield entry
4078 continuation_list[0] = self._extract_continuation(renderer)
4079 break
4080
4081 if not continuation_list[0]:
4082 continuation_list[0] = self._extract_continuation(is_renderer)
4083
4084 if not continuation_list[0]:
4085 continuation_list[0] = self._extract_continuation(parent_renderer)
4086
4087 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4088 continuation_list = [None]
4089 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4090 tab_content = try_get(tab, lambda x: x['content'], dict)
4091 if not tab_content:
4092 return
4093 parent_renderer = (
4094 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4095 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4096 yield from extract_entries(parent_renderer)
4097 continuation = continuation_list[0]
4098
4099 for page_num in itertools.count(1):
4100 if not continuation:
4101 break
4102 headers = self.generate_api_headers(
4103 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4104 response = self._extract_response(
4105 item_id=f'{item_id} page {page_num}',
4106 query=continuation, headers=headers, ytcfg=ytcfg,
4107 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4108
4109 if not response:
4110 break
4111 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4112 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4113 visitor_data = self._extract_visitor_data(response) or visitor_data
4114
4115 known_continuation_renderers = {
4116 'playlistVideoListContinuation': self._playlist_entries,
4117 'gridContinuation': self._grid_entries,
4118 'itemSectionContinuation': self._post_thread_continuation_entries,
4119 'sectionListContinuation': extract_entries, # for feeds
4120 }
4121 continuation_contents = try_get(
4122 response, lambda x: x['continuationContents'], dict) or {}
4123 continuation_renderer = None
4124 for key, value in continuation_contents.items():
4125 if key not in known_continuation_renderers:
4126 continue
4127 continuation_renderer = value
4128 continuation_list = [None]
4129 yield from known_continuation_renderers[key](continuation_renderer)
4130 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4131 break
4132 if continuation_renderer:
4133 continue
4134
4135 known_renderers = {
4136 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4137 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4138 'gridVideoRenderer': (self._grid_entries, 'items'),
4139 'gridChannelRenderer': (self._grid_entries, 'items'),
4140 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4141 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4142 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4143 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4144 }
4145 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4146 continuation_items = try_get(
4147 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4148 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4149 video_items_renderer = None
4150 for key, value in continuation_item.items():
4151 if key not in known_renderers:
4152 continue
4153 video_items_renderer = {known_renderers[key][1]: continuation_items}
4154 continuation_list = [None]
4155 yield from known_renderers[key][0](video_items_renderer)
4156 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4157 break
4158 if video_items_renderer:
4159 continue
4160 break
4161
4162 @staticmethod
4163 def _extract_selected_tab(tabs, fatal=True):
4164 for tab in tabs:
4165 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4166 if renderer.get('selected') is True:
4167 return renderer
4168 else:
4169 if fatal:
4170 raise ExtractorError('Unable to find selected tab')
4171
4172 def _extract_uploader(self, data):
4173 uploader = {}
4174 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4175 owner = try_get(
4176 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4177 if owner:
4178 owner_text = owner.get('text')
4179 uploader['uploader'] = self._search_regex(
4180 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4181 uploader['uploader_id'] = try_get(
4182 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4183 uploader['uploader_url'] = urljoin(
4184 'https://www.youtube.com/',
4185 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4186 return {k: v for k, v in uploader.items() if v is not None}
4187
4188 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4189 playlist_id = title = description = channel_url = channel_name = channel_id = None
4190 tags = []
4191
4192 selected_tab = self._extract_selected_tab(tabs)
4193 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4194 renderer = try_get(
4195 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4196 if renderer:
4197 channel_name = renderer.get('title')
4198 channel_url = renderer.get('channelUrl')
4199 channel_id = renderer.get('externalId')
4200 else:
4201 renderer = try_get(
4202 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4203
4204 if renderer:
4205 title = renderer.get('title')
4206 description = renderer.get('description', '')
4207 playlist_id = channel_id
4208 tags = renderer.get('keywords', '').split()
4209
4210 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4211 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4212 def _get_uncropped(url):
4213 return url_or_none((url or '').split('=')[0] + '=s0')
4214
4215 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4216 if avatar_thumbnails:
4217 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4218 if uncropped_avatar:
4219 avatar_thumbnails.append({
4220 'url': uncropped_avatar,
4221 'id': 'avatar_uncropped',
4222 'preference': 1
4223 })
4224
4225 channel_banners = self._extract_thumbnails(
4226 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4227 for banner in channel_banners:
4228 banner['preference'] = -10
4229
4230 if channel_banners:
4231 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4232 if uncropped_banner:
4233 channel_banners.append({
4234 'url': uncropped_banner,
4235 'id': 'banner_uncropped',
4236 'preference': -5
4237 })
4238
4239 primary_thumbnails = self._extract_thumbnails(
4240 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4241
4242 if playlist_id is None:
4243 playlist_id = item_id
4244
4245 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4246 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4247 if title is None:
4248 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4249 title += format_field(selected_tab, 'title', ' - %s')
4250 title += format_field(selected_tab, 'expandedText', ' - %s')
4251
4252 metadata = {
4253 'playlist_id': playlist_id,
4254 'playlist_title': title,
4255 'playlist_description': description,
4256 'uploader': channel_name,
4257 'uploader_id': channel_id,
4258 'uploader_url': channel_url,
4259 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4260 'tags': tags,
4261 'view_count': self._get_count(playlist_stats, 1),
4262 'availability': self._extract_availability(data),
4263 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4264 'playlist_count': self._get_count(playlist_stats, 0),
4265 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4266 }
4267 if not channel_id:
4268 metadata.update(self._extract_uploader(data))
4269 metadata.update({
4270 'channel': metadata['uploader'],
4271 'channel_id': metadata['uploader_id'],
4272 'channel_url': metadata['uploader_url']})
4273 return self.playlist_result(
4274 self._entries(
4275 selected_tab, playlist_id, ytcfg,
4276 self._extract_account_syncid(ytcfg, data),
4277 self._extract_visitor_data(data, ytcfg)),
4278 **metadata)
4279
4280 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4281 first_id = last_id = response = None
4282 for page_num in itertools.count(1):
4283 videos = list(self._playlist_entries(playlist))
4284 if not videos:
4285 return
4286 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4287 if start >= len(videos):
4288 return
4289 yield from videos[start:]
4290 first_id = first_id or videos[0]['id']
4291 last_id = videos[-1]['id']
4292 watch_endpoint = try_get(
4293 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4294 headers = self.generate_api_headers(
4295 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4296 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4297 query = {
4298 'playlistId': playlist_id,
4299 'videoId': watch_endpoint.get('videoId') or last_id,
4300 'index': watch_endpoint.get('index') or len(videos),
4301 'params': watch_endpoint.get('params') or 'OAE%3D'
4302 }
4303 response = self._extract_response(
4304 item_id='%s page %d' % (playlist_id, page_num),
4305 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4306 check_get_keys='contents'
4307 )
4308 playlist = try_get(
4309 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4310
4311 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4312 title = playlist.get('title') or try_get(
4313 data, lambda x: x['titleText']['simpleText'], compat_str)
4314 playlist_id = playlist.get('playlistId') or item_id
4315
4316 # Delegating everything except mix playlists to regular tab-based playlist URL
4317 playlist_url = urljoin(url, try_get(
4318 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4319 compat_str))
4320
4321 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4322 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4323 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4324
4325 if playlist_url and playlist_url != url and not is_known_unviewable:
4326 return self.url_result(
4327 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4328 video_title=title)
4329
4330 return self.playlist_result(
4331 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4332 playlist_id=playlist_id, playlist_title=title)
4333
4334 def _extract_availability(self, data):
4335 """
4336 Gets the availability of a given playlist/tab.
4337 Note: Unless YouTube tells us explicitly, we do not assume it is public
4338 @param data: response
4339 """
4340 is_private = is_unlisted = None
4341 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4342 badge_labels = self._extract_badges(renderer)
4343
4344 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4345 privacy_dropdown_entries = try_get(
4346 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4347 for renderer_dict in privacy_dropdown_entries:
4348 is_selected = try_get(
4349 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4350 if not is_selected:
4351 continue
4352 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4353 if label:
4354 badge_labels.add(label.lower())
4355 break
4356
4357 for badge_label in badge_labels:
4358 if badge_label == 'unlisted':
4359 is_unlisted = True
4360 elif badge_label == 'private':
4361 is_private = True
4362 elif badge_label == 'public':
4363 is_unlisted = is_private = False
4364 return self._availability(is_private, False, False, False, is_unlisted)
4365
4366 @staticmethod
4367 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4368 sidebar_renderer = try_get(
4369 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4370 for item in sidebar_renderer:
4371 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4372 if renderer:
4373 return renderer
4374
4375 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4376 """
4377 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4378 """
4379 browse_id = params = None
4380 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4381 if not renderer:
4382 return
4383 menu_renderer = try_get(
4384 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4385 for menu_item in menu_renderer:
4386 if not isinstance(menu_item, dict):
4387 continue
4388 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4389 text = try_get(
4390 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4391 if not text or text.lower() != 'show unavailable videos':
4392 continue
4393 browse_endpoint = try_get(
4394 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4395 browse_id = browse_endpoint.get('browseId')
4396 params = browse_endpoint.get('params')
4397 break
4398
4399 headers = self.generate_api_headers(
4400 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4401 visitor_data=self._extract_visitor_data(data, ytcfg))
4402 query = {
4403 'params': params or 'wgYCCAA=',
4404 'browseId': browse_id or 'VL%s' % item_id
4405 }
4406 return self._extract_response(
4407 item_id=item_id, headers=headers, query=query,
4408 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4409 note='Downloading API JSON with unavailable videos')
4410
4411 @functools.cached_property
4412 def skip_webpage(self):
4413 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4414
4415 def _extract_webpage(self, url, item_id, fatal=True):
4416 retries = self.get_param('extractor_retries', 3)
4417 count = -1
4418 webpage = data = last_error = None
4419 while count < retries:
4420 count += 1
4421 # Sometimes youtube returns a webpage with incomplete ytInitialData
4422 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4423 if last_error:
4424 self.report_warning('%s. Retrying ...' % last_error)
4425 try:
4426 webpage = self._download_webpage(
4427 url, item_id,
4428 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4429 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4430 except ExtractorError as e:
4431 if isinstance(e.cause, network_exceptions):
4432 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4433 last_error = error_to_compat_str(e.cause or e.msg)
4434 if count < retries:
4435 continue
4436 if fatal:
4437 raise
4438 self.report_warning(error_to_compat_str(e))
4439 break
4440 else:
4441 try:
4442 self._extract_and_report_alerts(data)
4443 except ExtractorError as e:
4444 if fatal:
4445 raise
4446 self.report_warning(error_to_compat_str(e))
4447 break
4448
4449 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4450 break
4451
4452 last_error = 'Incomplete yt initial data received'
4453 if count >= retries:
4454 if fatal:
4455 raise ExtractorError(last_error)
4456 self.report_warning(last_error)
4457 break
4458
4459 return webpage, data
4460
4461 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4462 """Use if failed to extract ytcfg (and data) from initial webpage"""
4463 if not ytcfg and self.is_authenticated:
4464 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4465 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4466 raise ExtractorError(
4467 f'{msg}. If you are not downloading private content, or '
4468 'your cookies are only for the first account and channel,'
4469 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4470 expected=True)
4471 self.report_warning(msg, only_once=True)
4472
4473 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4474 data = None
4475 if not self.skip_webpage:
4476 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4477 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4478 # Reject webpage data if redirected to home page without explicitly requesting
4479 selected_tab = self._extract_selected_tab(traverse_obj(
4480 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4481 if (url != 'https://www.youtube.com/feed/recommended'
4482 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4483 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4484 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4485 if fatal:
4486 raise ExtractorError(msg, expected=True)
4487 self.report_warning(msg, only_once=True)
4488 if not data:
4489 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4490 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4491 return data, ytcfg
4492
4493 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4494 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4495 resolve_response = self._extract_response(
4496 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4497 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4498 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4499 for ep_key, ep in endpoints.items():
4500 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4501 if params:
4502 return self._extract_response(
4503 item_id=item_id, query=params, ep=ep, headers=headers,
4504 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4505 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4506 err_note = 'Failed to resolve url (does the playlist exist?)'
4507 if fatal:
4508 raise ExtractorError(err_note, expected=True)
4509 self.report_warning(err_note, item_id)
4510
4511 _SEARCH_PARAMS = None
4512
4513 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4514 data = {'query': query}
4515 if params is NO_DEFAULT:
4516 params = self._SEARCH_PARAMS
4517 if params:
4518 data['params'] = params
4519
4520 content_keys = (
4521 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4522 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4523 # ytmusic search
4524 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4525 ('continuationContents', ),
4526 )
4527 display_id = f'query "{query}"'
4528 check_get_keys = tuple({keys[0] for keys in content_keys})
4529 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4530 self._report_playlist_authcheck(ytcfg, fatal=False)
4531
4532 continuation_list = [None]
4533 search = None
4534 for page_num in itertools.count(1):
4535 data.update(continuation_list[0] or {})
4536 headers = self.generate_api_headers(
4537 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4538 search = self._extract_response(
4539 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4540 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4541 slr_contents = traverse_obj(search, *content_keys)
4542 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4543 if not continuation_list[0]:
4544 break
4545
4546
4547class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4548 IE_DESC = 'YouTube Tabs'
4549 _VALID_URL = r'''(?x:
4550 https?://
4551 (?:\w+\.)?
4552 (?:
4553 youtube(?:kids)?\.com|
4554 %(invidious)s
4555 )/
4556 (?:
4557 (?P<channel_type>channel|c|user|browse)/|
4558 (?P<not_channel>
4559 feed/|hashtag/|
4560 (?:playlist|watch)\?.*?\blist=
4561 )|
4562 (?!(?:%(reserved_names)s)\b) # Direct URLs
4563 )
4564 (?P<id>[^/?\#&]+)
4565 )''' % {
4566 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4567 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4568 }
4569 IE_NAME = 'youtube:tab'
4570
4571 _TESTS = [{
4572 'note': 'playlists, multipage',
4573 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4574 'playlist_mincount': 94,
4575 'info_dict': {
4576 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4577 'title': 'Igor Kleiner - Playlists',
4578 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4579 'uploader': 'Igor Kleiner',
4580 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4581 'channel': 'Igor Kleiner',
4582 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4583 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4584 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4585 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4586 'channel_follower_count': int
4587 },
4588 }, {
4589 'note': 'playlists, multipage, different order',
4590 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4591 'playlist_mincount': 94,
4592 'info_dict': {
4593 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4594 'title': 'Igor Kleiner - Playlists',
4595 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4596 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4597 'uploader': 'Igor Kleiner',
4598 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4599 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4600 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4601 'channel': 'Igor Kleiner',
4602 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4603 'channel_follower_count': int
4604 },
4605 }, {
4606 'note': 'playlists, series',
4607 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4608 'playlist_mincount': 5,
4609 'info_dict': {
4610 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4611 'title': '3Blue1Brown - Playlists',
4612 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4613 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4614 'uploader': '3Blue1Brown',
4615 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4616 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4617 'channel': '3Blue1Brown',
4618 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4619 'tags': ['Mathematics'],
4620 'channel_follower_count': int
4621 },
4622 }, {
4623 'note': 'playlists, singlepage',
4624 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4625 'playlist_mincount': 4,
4626 'info_dict': {
4627 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4628 'title': 'ThirstForScience - Playlists',
4629 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4630 'uploader': 'ThirstForScience',
4631 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4632 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4633 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4634 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4635 'tags': 'count:13',
4636 'channel': 'ThirstForScience',
4637 'channel_follower_count': int
4638 }
4639 }, {
4640 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4641 'only_matching': True,
4642 }, {
4643 'note': 'basic, single video playlist',
4644 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4645 'info_dict': {
4646 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4647 'uploader': 'Sergey M.',
4648 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4649 'title': 'youtube-dl public playlist',
4650 'description': '',
4651 'tags': [],
4652 'view_count': int,
4653 'modified_date': '20201130',
4654 'channel': 'Sergey M.',
4655 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4656 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4657 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4658 },
4659 'playlist_count': 1,
4660 }, {
4661 'note': 'empty playlist',
4662 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4663 'info_dict': {
4664 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4665 'uploader': 'Sergey M.',
4666 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4667 'title': 'youtube-dl empty playlist',
4668 'tags': [],
4669 'channel': 'Sergey M.',
4670 'description': '',
4671 'modified_date': '20160902',
4672 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4673 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4674 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4675 },
4676 'playlist_count': 0,
4677 }, {
4678 'note': 'Home tab',
4679 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4680 'info_dict': {
4681 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4682 'title': 'lex will - Home',
4683 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4684 'uploader': 'lex will',
4685 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4686 'channel': 'lex will',
4687 'tags': ['bible', 'history', 'prophesy'],
4688 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4689 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4690 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4691 'channel_follower_count': int
4692 },
4693 'playlist_mincount': 2,
4694 }, {
4695 'note': 'Videos tab',
4696 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4697 'info_dict': {
4698 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4699 'title': 'lex will - Videos',
4700 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4701 'uploader': 'lex will',
4702 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4703 'tags': ['bible', 'history', 'prophesy'],
4704 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4705 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4706 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4707 'channel': 'lex will',
4708 'channel_follower_count': int
4709 },
4710 'playlist_mincount': 975,
4711 }, {
4712 'note': 'Videos tab, sorted by popular',
4713 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4714 'info_dict': {
4715 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4716 'title': 'lex will - Videos',
4717 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4718 'uploader': 'lex will',
4719 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4720 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4721 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4722 'channel': 'lex will',
4723 'tags': ['bible', 'history', 'prophesy'],
4724 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4725 'channel_follower_count': int
4726 },
4727 'playlist_mincount': 199,
4728 }, {
4729 'note': 'Playlists tab',
4730 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4731 'info_dict': {
4732 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4733 'title': 'lex will - Playlists',
4734 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4735 'uploader': 'lex will',
4736 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4737 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4738 'channel': 'lex will',
4739 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4740 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'tags': ['bible', 'history', 'prophesy'],
4742 'channel_follower_count': int
4743 },
4744 'playlist_mincount': 17,
4745 }, {
4746 'note': 'Community tab',
4747 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4748 'info_dict': {
4749 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4750 'title': 'lex will - Community',
4751 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4752 'uploader': 'lex will',
4753 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4754 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4755 'channel': 'lex will',
4756 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4757 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4758 'tags': ['bible', 'history', 'prophesy'],
4759 'channel_follower_count': int
4760 },
4761 'playlist_mincount': 18,
4762 }, {
4763 'note': 'Channels tab',
4764 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4765 'info_dict': {
4766 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4767 'title': 'lex will - Channels',
4768 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4769 'uploader': 'lex will',
4770 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4771 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4772 'channel': 'lex will',
4773 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4774 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4775 'tags': ['bible', 'history', 'prophesy'],
4776 'channel_follower_count': int
4777 },
4778 'playlist_mincount': 12,
4779 }, {
4780 'note': 'Search tab',
4781 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4782 'playlist_mincount': 40,
4783 'info_dict': {
4784 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4785 'title': '3Blue1Brown - Search - linear algebra',
4786 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4787 'uploader': '3Blue1Brown',
4788 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4789 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4790 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4791 'tags': ['Mathematics'],
4792 'channel': '3Blue1Brown',
4793 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4794 'channel_follower_count': int
4795 },
4796 }, {
4797 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4798 'only_matching': True,
4799 }, {
4800 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4801 'only_matching': True,
4802 }, {
4803 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4804 'only_matching': True,
4805 }, {
4806 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4807 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4808 'info_dict': {
4809 'title': '29C3: Not my department',
4810 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4811 'uploader': 'Christiaan008',
4812 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4813 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4814 'tags': [],
4815 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4816 'view_count': int,
4817 'modified_date': '20150605',
4818 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4819 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4820 'channel': 'Christiaan008',
4821 },
4822 'playlist_count': 96,
4823 }, {
4824 'note': 'Large playlist',
4825 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4826 'info_dict': {
4827 'title': 'Uploads from Cauchemar',
4828 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4829 'uploader': 'Cauchemar',
4830 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4831 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4832 'tags': [],
4833 'modified_date': r're:\d{8}',
4834 'channel': 'Cauchemar',
4835 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4836 'view_count': int,
4837 'description': '',
4838 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4839 },
4840 'playlist_mincount': 1123,
4841 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4842 }, {
4843 'note': 'even larger playlist, 8832 videos',
4844 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4845 'only_matching': True,
4846 }, {
4847 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4848 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4849 'info_dict': {
4850 'title': 'Uploads from Interstellar Movie',
4851 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4852 'uploader': 'Interstellar Movie',
4853 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4854 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4855 'tags': [],
4856 'view_count': int,
4857 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4858 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4859 'channel': 'Interstellar Movie',
4860 'description': '',
4861 'modified_date': r're:\d{8}',
4862 },
4863 'playlist_mincount': 21,
4864 }, {
4865 'note': 'Playlist with "show unavailable videos" button',
4866 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4867 'info_dict': {
4868 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4869 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4870 'uploader': 'Phim Siêu Nhân Nhật Bản',
4871 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4872 'view_count': int,
4873 'channel': 'Phim Siêu Nhân Nhật Bản',
4874 'tags': [],
4875 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4876 'description': '',
4877 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4878 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4879 'modified_date': r're:\d{8}',
4880 },
4881 'playlist_mincount': 200,
4882 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4883 }, {
4884 'note': 'Playlist with unavailable videos in page 7',
4885 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4886 'info_dict': {
4887 'title': 'Uploads from BlankTV',
4888 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4889 'uploader': 'BlankTV',
4890 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4891 'channel': 'BlankTV',
4892 'channel_url': 'https://www.youtube.com/c/blanktv',
4893 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4894 'view_count': int,
4895 'tags': [],
4896 'uploader_url': 'https://www.youtube.com/c/blanktv',
4897 'modified_date': r're:\d{8}',
4898 'description': '',
4899 },
4900 'playlist_mincount': 1000,
4901 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4902 }, {
4903 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4904 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4905 'info_dict': {
4906 'title': 'Data Analysis with Dr Mike Pound',
4907 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4908 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4909 'uploader': 'Computerphile',
4910 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4911 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4912 'tags': [],
4913 'view_count': int,
4914 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4915 'channel_url': 'https://www.youtube.com/user/Computerphile',
4916 'channel': 'Computerphile',
4917 },
4918 'playlist_mincount': 11,
4919 }, {
4920 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4921 'only_matching': True,
4922 }, {
4923 'note': 'Playlist URL that does not actually serve a playlist',
4924 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4925 'info_dict': {
4926 'id': 'FqZTN594JQw',
4927 'ext': 'webm',
4928 'title': "Smiley's People 01 detective, Adventure Series, Action",
4929 'uploader': 'STREEM',
4930 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4931 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4932 'upload_date': '20150526',
4933 'license': 'Standard YouTube License',
4934 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4935 'categories': ['People & Blogs'],
4936 'tags': list,
4937 'view_count': int,
4938 'like_count': int,
4939 },
4940 'params': {
4941 'skip_download': True,
4942 },
4943 'skip': 'This video is not available.',
4944 'add_ie': [YoutubeIE.ie_key()],
4945 }, {
4946 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4947 'only_matching': True,
4948 }, {
4949 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4950 'only_matching': True,
4951 }, {
4952 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4953 'info_dict': {
4954 'id': 'GgL890LIznQ', # This will keep changing
4955 'ext': 'mp4',
4956 'title': str,
4957 'uploader': 'Sky News',
4958 'uploader_id': 'skynews',
4959 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4960 'upload_date': r're:\d{8}',
4961 'description': str,
4962 'categories': ['News & Politics'],
4963 'tags': list,
4964 'like_count': int,
4965 'release_timestamp': 1642502819,
4966 'channel': 'Sky News',
4967 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4968 'age_limit': 0,
4969 'view_count': int,
4970 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4971 'playable_in_embed': True,
4972 'release_date': '20220118',
4973 'availability': 'public',
4974 'live_status': 'is_live',
4975 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4976 'channel_follower_count': int
4977 },
4978 'params': {
4979 'skip_download': True,
4980 },
4981 'expected_warnings': ['Ignoring subtitle tracks found in '],
4982 }, {
4983 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4984 'info_dict': {
4985 'id': 'a48o2S1cPoo',
4986 'ext': 'mp4',
4987 'title': 'The Young Turks - Live Main Show',
4988 'uploader': 'The Young Turks',
4989 'uploader_id': 'TheYoungTurks',
4990 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4991 'upload_date': '20150715',
4992 'license': 'Standard YouTube License',
4993 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4994 'categories': ['News & Politics'],
4995 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4996 'like_count': int,
4997 },
4998 'params': {
4999 'skip_download': True,
5000 },
5001 'only_matching': True,
5002 }, {
5003 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5004 'only_matching': True,
5005 }, {
5006 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5007 'only_matching': True,
5008 }, {
5009 'note': 'A channel that is not live. Should raise error',
5010 'url': 'https://www.youtube.com/user/numberphile/live',
5011 'only_matching': True,
5012 }, {
5013 'url': 'https://www.youtube.com/feed/trending',
5014 'only_matching': True,
5015 }, {
5016 'url': 'https://www.youtube.com/feed/library',
5017 'only_matching': True,
5018 }, {
5019 'url': 'https://www.youtube.com/feed/history',
5020 'only_matching': True,
5021 }, {
5022 'url': 'https://www.youtube.com/feed/subscriptions',
5023 'only_matching': True,
5024 }, {
5025 'url': 'https://www.youtube.com/feed/watch_later',
5026 'only_matching': True,
5027 }, {
5028 'note': 'Recommended - redirects to home page.',
5029 'url': 'https://www.youtube.com/feed/recommended',
5030 'only_matching': True,
5031 }, {
5032 'note': 'inline playlist with not always working continuations',
5033 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5034 'only_matching': True,
5035 }, {
5036 'url': 'https://www.youtube.com/course',
5037 'only_matching': True,
5038 }, {
5039 'url': 'https://www.youtube.com/zsecurity',
5040 'only_matching': True,
5041 }, {
5042 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5043 'only_matching': True,
5044 }, {
5045 'url': 'https://www.youtube.com/TheYoungTurks/live',
5046 'only_matching': True,
5047 }, {
5048 'url': 'https://www.youtube.com/hashtag/cctv9',
5049 'info_dict': {
5050 'id': 'cctv9',
5051 'title': '#cctv9',
5052 'tags': [],
5053 },
5054 'playlist_mincount': 350,
5055 }, {
5056 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5057 'only_matching': True,
5058 }, {
5059 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5060 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5061 'only_matching': True
5062 }, {
5063 'note': '/browse/ should redirect to /channel/',
5064 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5065 'only_matching': True
5066 }, {
5067 'note': 'VLPL, should redirect to playlist?list=PL...',
5068 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5069 'info_dict': {
5070 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5071 'uploader': 'NoCopyrightSounds',
5072 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5073 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5074 'title': 'NCS Releases',
5075 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5076 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5077 'modified_date': r're:\d{8}',
5078 'view_count': int,
5079 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5080 'tags': [],
5081 'channel': 'NoCopyrightSounds',
5082 },
5083 'playlist_mincount': 166,
5084 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5085 }, {
5086 'note': 'Topic, should redirect to playlist?list=UU...',
5087 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5088 'info_dict': {
5089 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5090 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5091 'title': 'Uploads from Royalty Free Music - Topic',
5092 'uploader': 'Royalty Free Music - Topic',
5093 'tags': [],
5094 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5095 'channel': 'Royalty Free Music - Topic',
5096 'view_count': int,
5097 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5098 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5099 'modified_date': r're:\d{8}',
5100 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5101 'description': '',
5102 },
5103 'expected_warnings': [
5104 'The URL does not have a videos tab',
5105 r'[Uu]navailable videos (are|will be) hidden',
5106 ],
5107 'playlist_mincount': 101,
5108 }, {
5109 'note': 'Topic without a UU playlist',
5110 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5111 'info_dict': {
5112 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5113 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5114 'tags': [],
5115 },
5116 'expected_warnings': [
5117 'the playlist redirect gave error',
5118 ],
5119 'playlist_mincount': 9,
5120 }, {
5121 'note': 'Youtube music Album',
5122 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5123 'info_dict': {
5124 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5125 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5126 'tags': [],
5127 'view_count': int,
5128 'description': '',
5129 'availability': 'unlisted',
5130 'modified_date': r're:\d{8}',
5131 },
5132 'playlist_count': 50,
5133 }, {
5134 'note': 'unlisted single video playlist',
5135 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5136 'info_dict': {
5137 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5138 'uploader': 'colethedj',
5139 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5140 'title': 'yt-dlp unlisted playlist test',
5141 'availability': 'unlisted',
5142 'tags': [],
5143 'modified_date': '20211208',
5144 'channel': 'colethedj',
5145 'view_count': int,
5146 'description': '',
5147 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5148 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5149 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5150 },
5151 'playlist_count': 1,
5152 }, {
5153 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5154 'url': 'https://www.youtube.com/feed/recommended',
5155 'info_dict': {
5156 'id': 'recommended',
5157 'title': 'recommended',
5158 'tags': [],
5159 },
5160 'playlist_mincount': 50,
5161 'params': {
5162 'skip_download': True,
5163 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5164 },
5165 }, {
5166 'note': 'API Fallback: /videos tab, sorted by oldest first',
5167 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5168 'info_dict': {
5169 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5170 'title': 'Cody\'sLab - Videos',
5171 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5172 'uploader': 'Cody\'sLab',
5173 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5174 'channel': 'Cody\'sLab',
5175 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5176 'tags': [],
5177 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5178 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5179 'channel_follower_count': int
5180 },
5181 'playlist_mincount': 650,
5182 'params': {
5183 'skip_download': True,
5184 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5185 },
5186 }, {
5187 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5188 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5189 'info_dict': {
5190 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5191 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5192 'title': 'Uploads from Royalty Free Music - Topic',
5193 'uploader': 'Royalty Free Music - Topic',
5194 'modified_date': r're:\d{8}',
5195 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5196 'description': '',
5197 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5198 'tags': [],
5199 'channel': 'Royalty Free Music - Topic',
5200 'view_count': int,
5201 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5202 },
5203 'expected_warnings': [
5204 'does not have a videos tab',
5205 r'[Uu]navailable videos (are|will be) hidden',
5206 ],
5207 'playlist_mincount': 101,
5208 'params': {
5209 'skip_download': True,
5210 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5211 },
5212 }, {
5213 'note': 'non-standard redirect to regional channel',
5214 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5215 'only_matching': True
5216 }, {
5217 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5218 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5219 'info_dict': {
5220 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5221 'modified_date': '20220407',
5222 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5223 'tags': [],
5224 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5225 'uploader': 'pukkandan',
5226 'availability': 'unlisted',
5227 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5228 'channel': 'pukkandan',
5229 'description': 'Test for collaborative playlist',
5230 'title': 'yt-dlp test - collaborative playlist',
5231 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5232 },
5233 'playlist_mincount': 2
5234 }]
5235
5236 @classmethod
5237 def suitable(cls, url):
5238 return False if YoutubeIE.suitable(url) else super().suitable(url)
5239
5240 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5241
5242 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5243 def _real_extract(self, url, smuggled_data):
5244 item_id = self._match_id(url)
5245 url = compat_urlparse.urlunparse(
5246 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5247 compat_opts = self.get_param('compat_opts', [])
5248
5249 def get_mobj(url):
5250 mobj = self._URL_RE.match(url).groupdict()
5251 mobj.update((k, '') for k, v in mobj.items() if v is None)
5252 return mobj
5253
5254 mobj, redirect_warning = get_mobj(url), None
5255 # Youtube returns incomplete data if tabname is not lower case
5256 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5257 if is_channel:
5258 if smuggled_data.get('is_music_url'):
5259 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5260 item_id = item_id[2:]
5261 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5262 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5263 mdata = self._extract_tab_endpoint(
5264 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5265 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5266 get_all=False, expected_type=compat_str)
5267 if not murl:
5268 raise ExtractorError('Failed to resolve album to playlist')
5269 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5270 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5271 pre = f'https://www.youtube.com/channel/{item_id}'
5272
5273 original_tab_name = tab
5274 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5275 # Home URLs should redirect to /videos/
5276 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5277 'To download only the videos in the home page, add a "/featured" to the URL')
5278 tab = '/videos'
5279
5280 url = ''.join((pre, tab, post))
5281 mobj = get_mobj(url)
5282
5283 # Handle both video/playlist URLs
5284 qs = parse_qs(url)
5285 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5286
5287 if not video_id and mobj['not_channel'].startswith('watch'):
5288 if not playlist_id:
5289 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5290 raise ExtractorError('Unable to recognize tab page')
5291 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5292 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5293 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5294 mobj = get_mobj(url)
5295
5296 if video_id and playlist_id:
5297 if self.get_param('noplaylist'):
5298 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5299 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5300 ie=YoutubeIE.ie_key(), video_id=video_id)
5301 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5302
5303 data, ytcfg = self._extract_data(url, item_id)
5304
5305 # YouTube may provide a non-standard redirect to the regional channel
5306 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5307 redirect_url = traverse_obj(
5308 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5309 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5310 redirect_url = ''.join((
5311 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5312 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5313 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5314
5315 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5316 if tabs:
5317 selected_tab = self._extract_selected_tab(tabs)
5318 selected_tab_name = selected_tab.get('title', '').lower()
5319 if selected_tab_name == 'home':
5320 selected_tab_name = 'featured'
5321 requested_tab_name = mobj['tab'][1:]
5322 if 'no-youtube-channel-redirect' not in compat_opts:
5323 if requested_tab_name == 'live':
5324 # Live tab should have redirected to the video
5325 raise ExtractorError('The channel is not currently live', expected=True)
5326 if requested_tab_name not in ('', selected_tab_name):
5327 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5328 if not original_tab_name:
5329 if item_id[:2] == 'UC':
5330 # Topic channels don't have /videos. Use the equivalent playlist instead
5331 pl_id = f'UU{item_id[2:]}'
5332 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5333 try:
5334 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5335 except ExtractorError:
5336 redirect_warning += ' and the playlist redirect gave error'
5337 else:
5338 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5339 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5340 if selected_tab_name and selected_tab_name != requested_tab_name:
5341 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5342 else:
5343 raise ExtractorError(redirect_warning, expected=True)
5344
5345 if redirect_warning:
5346 self.to_screen(redirect_warning)
5347 self.write_debug(f'Final URL: {url}')
5348
5349 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5350 if 'no-youtube-unavailable-videos' not in compat_opts:
5351 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5352 self._extract_and_report_alerts(data, only_once=True)
5353 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5354 if tabs:
5355 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5356
5357 playlist = traverse_obj(
5358 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5359 if playlist:
5360 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5361
5362 video_id = traverse_obj(
5363 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5364 if video_id:
5365 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5366 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5367 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5368 ie=YoutubeIE.ie_key(), video_id=video_id)
5369
5370 raise ExtractorError('Unable to recognize tab page')
5371
5372
5373class YoutubePlaylistIE(InfoExtractor):
5374 IE_DESC = 'YouTube playlists'
5375 _VALID_URL = r'''(?x)(?:
5376 (?:https?://)?
5377 (?:\w+\.)?
5378 (?:
5379 (?:
5380 youtube(?:kids)?\.com|
5381 %(invidious)s
5382 )
5383 /.*?\?.*?\blist=
5384 )?
5385 (?P<id>%(playlist_id)s)
5386 )''' % {
5387 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5388 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5389 }
5390 IE_NAME = 'youtube:playlist'
5391 _TESTS = [{
5392 'note': 'issue #673',
5393 'url': 'PLBB231211A4F62143',
5394 'info_dict': {
5395 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5396 'id': 'PLBB231211A4F62143',
5397 'uploader': 'Wickman',
5398 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5399 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5400 'view_count': int,
5401 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5402 'modified_date': r're:\d{8}',
5403 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5404 'channel': 'Wickman',
5405 'tags': [],
5406 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5407 },
5408 'playlist_mincount': 29,
5409 }, {
5410 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5411 'info_dict': {
5412 'title': 'YDL_safe_search',
5413 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5414 },
5415 'playlist_count': 2,
5416 'skip': 'This playlist is private',
5417 }, {
5418 'note': 'embedded',
5419 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5420 'playlist_count': 4,
5421 'info_dict': {
5422 'title': 'JODA15',
5423 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5424 'uploader': 'milan',
5425 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5426 'description': '',
5427 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5428 'tags': [],
5429 'modified_date': '20140919',
5430 'view_count': int,
5431 'channel': 'milan',
5432 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5433 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5434 },
5435 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5436 }, {
5437 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5438 'playlist_mincount': 654,
5439 'info_dict': {
5440 'title': '2018 Chinese New Singles (11/6 updated)',
5441 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5442 'uploader': 'LBK',
5443 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5444 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5445 'channel': 'LBK',
5446 'view_count': int,
5447 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5448 'tags': [],
5449 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5450 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5451 'modified_date': r're:\d{8}',
5452 },
5453 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5454 }, {
5455 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5456 'only_matching': True,
5457 }, {
5458 # music album playlist
5459 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5460 'only_matching': True,
5461 }]
5462
5463 @classmethod
5464 def suitable(cls, url):
5465 if YoutubeTabIE.suitable(url):
5466 return False
5467 from ..utils import parse_qs
5468 qs = parse_qs(url)
5469 if qs.get('v', [None])[0]:
5470 return False
5471 return super().suitable(url)
5472
5473 def _real_extract(self, url):
5474 playlist_id = self._match_id(url)
5475 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5476 url = update_url_query(
5477 'https://www.youtube.com/playlist',
5478 parse_qs(url) or {'list': playlist_id})
5479 if is_music_url:
5480 url = smuggle_url(url, {'is_music_url': True})
5481 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5482
5483
5484class YoutubeYtBeIE(InfoExtractor):
5485 IE_DESC = 'youtu.be'
5486 _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}
5487 _TESTS = [{
5488 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5489 'info_dict': {
5490 'id': 'yeWKywCrFtk',
5491 'ext': 'mp4',
5492 'title': 'Small Scale Baler and Braiding Rugs',
5493 'uploader': 'Backus-Page House Museum',
5494 'uploader_id': 'backuspagemuseum',
5495 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5496 'upload_date': '20161008',
5497 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5498 'categories': ['Nonprofits & Activism'],
5499 'tags': list,
5500 'like_count': int,
5501 'age_limit': 0,
5502 'playable_in_embed': True,
5503 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5504 'channel': 'Backus-Page House Museum',
5505 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5506 'live_status': 'not_live',
5507 'view_count': int,
5508 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5509 'availability': 'public',
5510 'duration': 59,
5511 },
5512 'params': {
5513 'noplaylist': True,
5514 'skip_download': True,
5515 },
5516 }, {
5517 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5518 'only_matching': True,
5519 }]
5520
5521 def _real_extract(self, url):
5522 mobj = self._match_valid_url(url)
5523 video_id = mobj.group('id')
5524 playlist_id = mobj.group('playlist_id')
5525 return self.url_result(
5526 update_url_query('https://www.youtube.com/watch', {
5527 'v': video_id,
5528 'list': playlist_id,
5529 'feature': 'youtu.be',
5530 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5531
5532
5533class YoutubeLivestreamEmbedIE(InfoExtractor):
5534 IE_DESC = 'YouTube livestream embeds'
5535 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5536 _TESTS = [{
5537 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5538 'only_matching': True,
5539 }]
5540
5541 def _real_extract(self, url):
5542 channel_id = self._match_id(url)
5543 return self.url_result(
5544 f'https://www.youtube.com/channel/{channel_id}/live',
5545 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5546
5547
5548class YoutubeYtUserIE(InfoExtractor):
5549 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5550 IE_NAME = 'youtube:user'
5551 _VALID_URL = r'ytuser:(?P<id>.+)'
5552 _TESTS = [{
5553 'url': 'ytuser:phihag',
5554 'only_matching': True,
5555 }]
5556
5557 def _real_extract(self, url):
5558 user_id = self._match_id(url)
5559 return self.url_result(
5560 'https://www.youtube.com/user/%s/videos' % user_id,
5561 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5562
5563
5564class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5565 IE_NAME = 'youtube:favorites'
5566 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5567 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5568 _LOGIN_REQUIRED = True
5569 _TESTS = [{
5570 'url': ':ytfav',
5571 'only_matching': True,
5572 }, {
5573 'url': ':ytfavorites',
5574 'only_matching': True,
5575 }]
5576
5577 def _real_extract(self, url):
5578 return self.url_result(
5579 'https://www.youtube.com/playlist?list=LL',
5580 ie=YoutubeTabIE.ie_key())
5581
5582
5583class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5584 IE_NAME = 'youtube:notif'
5585 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5586 _VALID_URL = r':ytnotif(?:ication)?s?'
5587 _LOGIN_REQUIRED = True
5588 _TESTS = [{
5589 'url': ':ytnotif',
5590 'only_matching': True,
5591 }, {
5592 'url': ':ytnotifications',
5593 'only_matching': True,
5594 }]
5595
5596 def _extract_notification_menu(self, response, continuation_list):
5597 notification_list = traverse_obj(
5598 response,
5599 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5600 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5601 expected_type=list) or []
5602 continuation_list[0] = None
5603 for item in notification_list:
5604 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5605 if entry:
5606 yield entry
5607 continuation = item.get('continuationItemRenderer')
5608 if continuation:
5609 continuation_list[0] = continuation
5610
5611 def _extract_notification_renderer(self, notification):
5612 video_id = traverse_obj(
5613 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5614 url = f'https://www.youtube.com/watch?v={video_id}'
5615 channel_id = None
5616 if not video_id:
5617 browse_ep = traverse_obj(
5618 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5619 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5620 post_id = self._search_regex(
5621 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5622 'post id', default=None)
5623 if not channel_id or not post_id:
5624 return
5625 # The direct /post url redirects to this in the browser
5626 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5627
5628 channel = traverse_obj(
5629 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5630 expected_type=str)
5631 title = self._search_regex(
5632 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5633 'video title', default=None)
5634 if title:
5635 title = title.replace('\xad', '') # remove soft hyphens
5636 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5637 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5638 else None)
5639 return {
5640 '_type': 'url',
5641 'url': url,
5642 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5643 'video_id': video_id,
5644 'title': title,
5645 'channel_id': channel_id,
5646 'channel': channel,
5647 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5648 'upload_date': upload_date,
5649 }
5650
5651 def _notification_menu_entries(self, ytcfg):
5652 continuation_list = [None]
5653 response = None
5654 for page in itertools.count(1):
5655 ctoken = traverse_obj(
5656 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5657 response = self._extract_response(
5658 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5659 ep='notification/get_notification_menu', check_get_keys='actions',
5660 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5661 yield from self._extract_notification_menu(response, continuation_list)
5662 if not continuation_list[0]:
5663 break
5664
5665 def _real_extract(self, url):
5666 display_id = 'notifications'
5667 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5668 self._report_playlist_authcheck(ytcfg)
5669 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5670
5671
5672class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5673 IE_DESC = 'YouTube search'
5674 IE_NAME = 'youtube:search'
5675 _SEARCH_KEY = 'ytsearch'
5676 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5677 _TESTS = [{
5678 'url': 'ytsearch5:youtube-dl test video',
5679 'playlist_count': 5,
5680 'info_dict': {
5681 'id': 'youtube-dl test video',
5682 'title': 'youtube-dl test video',
5683 }
5684 }]
5685
5686
5687class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5688 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5689 _SEARCH_KEY = 'ytsearchdate'
5690 IE_DESC = 'YouTube search, newest videos first'
5691 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5692 _TESTS = [{
5693 'url': 'ytsearchdate5:youtube-dl test video',
5694 'playlist_count': 5,
5695 'info_dict': {
5696 'id': 'youtube-dl test video',
5697 'title': 'youtube-dl test video',
5698 }
5699 }]
5700
5701
5702class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5703 IE_DESC = 'YouTube search URLs with sorting and filter support'
5704 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5705 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5706 _TESTS = [{
5707 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5708 'playlist_mincount': 5,
5709 'info_dict': {
5710 'id': 'youtube-dl test video',
5711 'title': 'youtube-dl test video',
5712 }
5713 }, {
5714 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5715 'playlist_mincount': 5,
5716 'info_dict': {
5717 'id': 'python',
5718 'title': 'python',
5719 }
5720 }, {
5721 'url': 'https://www.youtube.com/results?search_query=%23cats',
5722 'playlist_mincount': 1,
5723 'info_dict': {
5724 'id': '#cats',
5725 'title': '#cats',
5726 'entries': [{
5727 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5728 'title': '#cats',
5729 }],
5730 },
5731 }, {
5732 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5733 'only_matching': True,
5734 }]
5735
5736 def _real_extract(self, url):
5737 qs = parse_qs(url)
5738 query = (qs.get('search_query') or qs.get('q'))[0]
5739 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5740
5741
5742class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5743 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5744 IE_NAME = 'youtube:music:search_url'
5745 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5746 _TESTS = [{
5747 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5748 'playlist_count': 16,
5749 'info_dict': {
5750 'id': 'royalty free music',
5751 'title': 'royalty free music',
5752 }
5753 }, {
5754 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5755 'playlist_mincount': 30,
5756 'info_dict': {
5757 'id': 'royalty free music - songs',
5758 'title': 'royalty free music - songs',
5759 },
5760 'params': {'extract_flat': 'in_playlist'}
5761 }, {
5762 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5763 'playlist_mincount': 30,
5764 'info_dict': {
5765 'id': 'royalty free music - community playlists',
5766 'title': 'royalty free music - community playlists',
5767 },
5768 'params': {'extract_flat': 'in_playlist'}
5769 }]
5770
5771 _SECTIONS = {
5772 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5773 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5774 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5775 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5776 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5777 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5778 }
5779
5780 def _real_extract(self, url):
5781 qs = parse_qs(url)
5782 query = (qs.get('search_query') or qs.get('q'))[0]
5783 params = qs.get('sp', (None,))[0]
5784 if params:
5785 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5786 else:
5787 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5788 params = self._SECTIONS.get(section)
5789 if not params:
5790 section = None
5791 title = join_nonempty(query, section, delim=' - ')
5792 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5793
5794
5795class YoutubeFeedsInfoExtractor(InfoExtractor):
5796 """
5797 Base class for feed extractors
5798 Subclasses must re-define the _FEED_NAME property.
5799 """
5800 _LOGIN_REQUIRED = True
5801 _FEED_NAME = 'feeds'
5802
5803 def _real_initialize(self):
5804 YoutubeBaseInfoExtractor._check_login_required(self)
5805
5806 @classproperty
5807 def IE_NAME(self):
5808 return f'youtube:{self._FEED_NAME}'
5809
5810 def _real_extract(self, url):
5811 return self.url_result(
5812 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5813
5814
5815class YoutubeWatchLaterIE(InfoExtractor):
5816 IE_NAME = 'youtube:watchlater'
5817 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5818 _VALID_URL = r':ytwatchlater'
5819 _TESTS = [{
5820 'url': ':ytwatchlater',
5821 'only_matching': True,
5822 }]
5823
5824 def _real_extract(self, url):
5825 return self.url_result(
5826 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5827
5828
5829class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5830 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5831 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5832 _FEED_NAME = 'recommended'
5833 _LOGIN_REQUIRED = False
5834 _TESTS = [{
5835 'url': ':ytrec',
5836 'only_matching': True,
5837 }, {
5838 'url': ':ytrecommended',
5839 'only_matching': True,
5840 }, {
5841 'url': 'https://youtube.com',
5842 'only_matching': True,
5843 }]
5844
5845
5846class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5847 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5848 _VALID_URL = r':ytsub(?:scription)?s?'
5849 _FEED_NAME = 'subscriptions'
5850 _TESTS = [{
5851 'url': ':ytsubs',
5852 'only_matching': True,
5853 }, {
5854 'url': ':ytsubscriptions',
5855 'only_matching': True,
5856 }]
5857
5858
5859class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5860 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5861 _VALID_URL = r':ythis(?:tory)?'
5862 _FEED_NAME = 'history'
5863 _TESTS = [{
5864 'url': ':ythistory',
5865 'only_matching': True,
5866 }]
5867
5868
5869class YoutubeStoriesIE(InfoExtractor):
5870 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5871 IE_NAME = 'youtube:stories'
5872 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5873 _TESTS = [{
5874 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5875 'only_matching': True,
5876 }]
5877
5878 def _real_extract(self, url):
5879 playlist_id = f'RLTD{self._match_id(url)}'
5880 return self.url_result(
5881 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5882 ie=YoutubeTabIE, video_id=playlist_id)
5883
5884
5885class YoutubeTruncatedURLIE(InfoExtractor):
5886 IE_NAME = 'youtube:truncated_url'
5887 IE_DESC = False # Do not list
5888 _VALID_URL = r'''(?x)
5889 (?:https?://)?
5890 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5891 (?:watch\?(?:
5892 feature=[a-z_]+|
5893 annotation_id=annotation_[^&]+|
5894 x-yt-cl=[0-9]+|
5895 hl=[^&]*|
5896 t=[0-9]+
5897 )?
5898 |
5899 attribution_link\?a=[^&]+
5900 )
5901 $
5902 '''
5903
5904 _TESTS = [{
5905 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5906 'only_matching': True,
5907 }, {
5908 'url': 'https://www.youtube.com/watch?',
5909 'only_matching': True,
5910 }, {
5911 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5912 'only_matching': True,
5913 }, {
5914 'url': 'https://www.youtube.com/watch?feature=foo',
5915 'only_matching': True,
5916 }, {
5917 'url': 'https://www.youtube.com/watch?hl=en-GB',
5918 'only_matching': True,
5919 }, {
5920 'url': 'https://www.youtube.com/watch?t=2372',
5921 'only_matching': True,
5922 }]
5923
5924 def _real_extract(self, url):
5925 raise ExtractorError(
5926 'Did you forget to quote the URL? Remember that & is a meta '
5927 'character in most shells, so you want to put the URL in quotes, '
5928 'like youtube-dl '
5929 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5930 ' or simply youtube-dl BaW_jenozKc .',
5931 expected=True)
5932
5933
5934class YoutubeClipIE(InfoExtractor):
5935 IE_NAME = 'youtube:clip'
5936 IE_DESC = False # Do not list
5937 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5938
5939 def _real_extract(self, url):
5940 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5941 return self.url_result(url, 'Generic')
5942
5943
5944class YoutubeTruncatedIDIE(InfoExtractor):
5945 IE_NAME = 'youtube:truncated_id'
5946 IE_DESC = False # Do not list
5947 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5948
5949 _TESTS = [{
5950 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5951 'only_matching': True,
5952 }]
5953
5954 def _real_extract(self, url):
5955 video_id = self._match_id(url)
5956 raise ExtractorError(
5957 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5958 expected=True)