]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[cookies] Detect profiles for cygwin/BSD (#3975)
[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 # isort: split
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*='
401 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
402
403 def _get_default_ytcfg(self, client='web'):
404 return copy.deepcopy(INNERTUBE_CLIENTS[client])
405
406 def _get_innertube_host(self, client='web'):
407 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
408
409 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
410 # try_get but with fallback to default ytcfg client values when present
411 _func = lambda y: try_get(y, getter, expected_type)
412 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
413
414 def _extract_client_name(self, ytcfg, default_client='web'):
415 return self._ytcfg_get_safe(
416 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
417 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
418
419 def _extract_client_version(self, ytcfg, default_client='web'):
420 return self._ytcfg_get_safe(
421 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
422 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
423
424 def _extract_api_key(self, ytcfg=None, default_client='web'):
425 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
426
427 def _extract_context(self, ytcfg=None, default_client='web'):
428 context = get_first(
429 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
430 # Enforce language and tz for extraction
431 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
432 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
433 return context
434
435 _SAPISID = None
436
437 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
438 time_now = round(time.time())
439 if self._SAPISID is None:
440 yt_cookies = self._get_cookies('https://www.youtube.com')
441 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
442 # See: https://github.com/yt-dlp/yt-dlp/issues/393
443 sapisid_cookie = dict_get(
444 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
445 if sapisid_cookie and sapisid_cookie.value:
446 self._SAPISID = sapisid_cookie.value
447 self.write_debug('Extracted SAPISID cookie')
448 # SAPISID cookie is required if not already present
449 if not yt_cookies.get('SAPISID'):
450 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
451 self._set_cookie(
452 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
453 else:
454 self._SAPISID = False
455 if not self._SAPISID:
456 return None
457 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
458 sapisidhash = hashlib.sha1(
459 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
460 return f'SAPISIDHASH {time_now}_{sapisidhash}'
461
462 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
463 note='Downloading API JSON', errnote='Unable to download API page',
464 context=None, api_key=None, api_hostname=None, default_client='web'):
465
466 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
467 data.update(query)
468 real_headers = self.generate_api_headers(default_client=default_client)
469 real_headers.update({'content-type': 'application/json'})
470 if headers:
471 real_headers.update(headers)
472 return self._download_json(
473 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
474 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
475 data=json.dumps(data).encode('utf8'), headers=real_headers,
476 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
477
478 def extract_yt_initial_data(self, item_id, webpage):
479 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=True)
480
481 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
482 return self._parse_json(self._search_regex(
483 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
484 regex), webpage, name, default='{}'), video_id, fatal=False, lenient=True)
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=vv8qTUWmulI',
2216 'info_dict': {
2217 'id': 'vv8qTUWmulI',
2218 'ext': 'mp4',
2219 'availability': 'unlisted',
2220 'view_count': int,
2221 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2222 'upload_date': '20220526',
2223 'categories': ['Education'],
2224 'title': 'Story',
2225 'channel': 'IT\'S HISTORY',
2226 'description': '',
2227 'uploader_id': 'BlastfromthePast',
2228 'duration': 12,
2229 'uploader': 'IT\'S HISTORY',
2230 'playable_in_embed': True,
2231 'age_limit': 0,
2232 'live_status': 'not_live',
2233 'tags': [],
2234 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2235 'uploader_url': 'http://www.youtube.com/user/BlastfromthePast',
2236 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2237 }
2238 }, {
2239 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2240 'info_dict': {
2241 'id': 'tjjjtzRLHvA',
2242 'ext': 'mp4',
2243 'title': 'ハッシュタグ無し };if window.ytcsi',
2244 'upload_date': '20220323',
2245 'like_count': int,
2246 'availability': 'unlisted',
2247 'channel': 'nao20010128nao',
2248 'thumbnail': 'https://i.ytimg.com/vi_webp/tjjjtzRLHvA/maxresdefault.webp',
2249 'age_limit': 0,
2250 'uploader': 'nao20010128nao',
2251 'uploader_id': 'nao20010128nao',
2252 'categories': ['Music'],
2253 'view_count': int,
2254 'description': '',
2255 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2256 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2257 'live_status': 'not_live',
2258 'playable_in_embed': True,
2259 'channel_follower_count': int,
2260 'duration': 6,
2261 'tags': [],
2262 'uploader_url': 'http://www.youtube.com/user/nao20010128nao',
2263 }
2264 }
2265 ]
2266
2267 @classmethod
2268 def suitable(cls, url):
2269 from ..utils import parse_qs
2270
2271 qs = parse_qs(url)
2272 if qs.get('list', [None])[0]:
2273 return False
2274 return super().suitable(url)
2275
2276 def __init__(self, *args, **kwargs):
2277 super().__init__(*args, **kwargs)
2278 self._code_cache = {}
2279 self._player_cache = {}
2280
2281 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2282 lock = threading.Lock()
2283
2284 is_live = True
2285 start_time = time.time()
2286 formats = [f for f in formats if f.get('is_from_start')]
2287
2288 def refetch_manifest(format_id, delay):
2289 nonlocal formats, start_time, is_live
2290 if time.time() <= start_time + delay:
2291 return
2292
2293 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2294 video_details = traverse_obj(
2295 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2296 microformats = traverse_obj(
2297 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2298 expected_type=dict, default=[])
2299 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2300 start_time = time.time()
2301
2302 def mpd_feed(format_id, delay):
2303 """
2304 @returns (manifest_url, manifest_stream_number, is_live) or None
2305 """
2306 with lock:
2307 refetch_manifest(format_id, delay)
2308
2309 f = next((f for f in formats if f['format_id'] == format_id), None)
2310 if not f:
2311 if not is_live:
2312 self.to_screen(f'{video_id}: Video is no longer live')
2313 else:
2314 self.report_warning(
2315 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2316 return None
2317 return f['manifest_url'], f['manifest_stream_number'], is_live
2318
2319 for f in formats:
2320 f['is_live'] = True
2321 f['protocol'] = 'http_dash_segments_generator'
2322 f['fragments'] = functools.partial(
2323 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2324
2325 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2326 FETCH_SPAN, MAX_DURATION = 5, 432000
2327
2328 mpd_url, stream_number, is_live = None, None, True
2329
2330 begin_index = 0
2331 download_start_time = ctx.get('start') or time.time()
2332
2333 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2334 if lack_early_segments:
2335 self.report_warning(bug_reports_message(
2336 'Starting download from the last 120 hours of the live stream since '
2337 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2338 lack_early_segments = True
2339
2340 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2341 fragments, fragment_base_url = None, None
2342
2343 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2344 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2345 # Obtain from MPD's maximum seq value
2346 old_mpd_url = mpd_url
2347 last_error = ctx.pop('last_error', None)
2348 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2349 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2350 or (mpd_url, stream_number, False))
2351 if not refresh_sequence:
2352 if expire_fast and not is_live:
2353 return False, last_seq
2354 elif old_mpd_url == mpd_url:
2355 return True, last_seq
2356 try:
2357 fmts, _ = self._extract_mpd_formats_and_subtitles(
2358 mpd_url, None, note=False, errnote=False, fatal=False)
2359 except ExtractorError:
2360 fmts = None
2361 if not fmts:
2362 no_fragment_score += 2
2363 return False, last_seq
2364 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2365 fragments = fmt_info['fragments']
2366 fragment_base_url = fmt_info['fragment_base_url']
2367 assert fragment_base_url
2368
2369 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2370 return True, _last_seq
2371
2372 while is_live:
2373 fetch_time = time.time()
2374 if no_fragment_score > 30:
2375 return
2376 if last_segment_url:
2377 # Obtain from "X-Head-Seqnum" header value from each segment
2378 try:
2379 urlh = self._request_webpage(
2380 last_segment_url, None, note=False, errnote=False, fatal=False)
2381 except ExtractorError:
2382 urlh = None
2383 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2384 if last_seq is None:
2385 no_fragment_score += 2
2386 last_segment_url = None
2387 continue
2388 else:
2389 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2390 no_fragment_score += 2
2391 if not should_continue:
2392 continue
2393
2394 if known_idx > last_seq:
2395 last_segment_url = None
2396 continue
2397
2398 last_seq += 1
2399
2400 if begin_index < 0 and known_idx < 0:
2401 # skip from the start when it's negative value
2402 known_idx = last_seq + begin_index
2403 if lack_early_segments:
2404 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2405 try:
2406 for idx in range(known_idx, last_seq):
2407 # do not update sequence here or you'll get skipped some part of it
2408 should_continue, _ = _extract_sequence_from_mpd(False, False)
2409 if not should_continue:
2410 known_idx = idx - 1
2411 raise ExtractorError('breaking out of outer loop')
2412 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2413 yield {
2414 'url': last_segment_url,
2415 }
2416 if known_idx == last_seq:
2417 no_fragment_score += 5
2418 else:
2419 no_fragment_score = 0
2420 known_idx = last_seq
2421 except ExtractorError:
2422 continue
2423
2424 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2425
2426 def _extract_player_url(self, *ytcfgs, webpage=None):
2427 player_url = traverse_obj(
2428 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2429 get_all=False, expected_type=compat_str)
2430 if not player_url:
2431 return
2432 return urljoin('https://www.youtube.com', player_url)
2433
2434 def _download_player_url(self, video_id, fatal=False):
2435 res = self._download_webpage(
2436 'https://www.youtube.com/iframe_api',
2437 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2438 if res:
2439 player_version = self._search_regex(
2440 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2441 if player_version:
2442 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2443
2444 def _signature_cache_id(self, example_sig):
2445 """ Return a string representation of a signature """
2446 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2447
2448 @classmethod
2449 def _extract_player_info(cls, player_url):
2450 for player_re in cls._PLAYER_INFO_RE:
2451 id_m = re.search(player_re, player_url)
2452 if id_m:
2453 break
2454 else:
2455 raise ExtractorError('Cannot identify player %r' % player_url)
2456 return id_m.group('id')
2457
2458 def _load_player(self, video_id, player_url, fatal=True):
2459 player_id = self._extract_player_info(player_url)
2460 if player_id not in self._code_cache:
2461 code = self._download_webpage(
2462 player_url, video_id, fatal=fatal,
2463 note='Downloading player ' + player_id,
2464 errnote='Download of %s failed' % player_url)
2465 if code:
2466 self._code_cache[player_id] = code
2467 return self._code_cache.get(player_id)
2468
2469 def _extract_signature_function(self, video_id, player_url, example_sig):
2470 player_id = self._extract_player_info(player_url)
2471
2472 # Read from filesystem cache
2473 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2474 assert os.path.basename(func_id) == func_id
2475
2476 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2477 if cache_spec is not None:
2478 return lambda s: ''.join(s[i] for i in cache_spec)
2479
2480 code = self._load_player(video_id, player_url)
2481 if code:
2482 res = self._parse_sig_js(code)
2483
2484 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2485 cache_res = res(test_string)
2486 cache_spec = [ord(c) for c in cache_res]
2487
2488 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2489 return res
2490
2491 def _print_sig_code(self, func, example_sig):
2492 if not self.get_param('youtube_print_sig_code'):
2493 return
2494
2495 def gen_sig_code(idxs):
2496 def _genslice(start, end, step):
2497 starts = '' if start == 0 else str(start)
2498 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2499 steps = '' if step == 1 else (':%d' % step)
2500 return f's[{starts}{ends}{steps}]'
2501
2502 step = None
2503 # Quelch pyflakes warnings - start will be set when step is set
2504 start = '(Never used)'
2505 for i, prev in zip(idxs[1:], idxs[:-1]):
2506 if step is not None:
2507 if i - prev == step:
2508 continue
2509 yield _genslice(start, prev, step)
2510 step = None
2511 continue
2512 if i - prev in [-1, 1]:
2513 step = i - prev
2514 start = prev
2515 continue
2516 else:
2517 yield 's[%d]' % prev
2518 if step is None:
2519 yield 's[%d]' % i
2520 else:
2521 yield _genslice(start, i, step)
2522
2523 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2524 cache_res = func(test_string)
2525 cache_spec = [ord(c) for c in cache_res]
2526 expr_code = ' + '.join(gen_sig_code(cache_spec))
2527 signature_id_tuple = '(%s)' % (
2528 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2529 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2530 ' return %s\n') % (signature_id_tuple, expr_code)
2531 self.to_screen('Extracted signature function:\n' + code)
2532
2533 def _parse_sig_js(self, jscode):
2534 funcname = self._search_regex(
2535 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2536 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2537 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2538 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2539 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}\(a,\d+\)',
2540 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\(\s*""\s*\)',
2541 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2542 # Obsolete patterns
2543 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2544 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2545 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2546 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2547 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2548 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2549 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2550 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2551 jscode, 'Initial JS player signature function name', group='sig')
2552
2553 jsi = JSInterpreter(jscode)
2554 initial_function = jsi.extract_function(funcname)
2555 return lambda s: initial_function([s])
2556
2557 def _decrypt_signature(self, s, video_id, player_url):
2558 """Turn the encrypted s field into a working signature"""
2559 try:
2560 player_id = (player_url, self._signature_cache_id(s))
2561 if player_id not in self._player_cache:
2562 func = self._extract_signature_function(video_id, player_url, s)
2563 self._player_cache[player_id] = func
2564 func = self._player_cache[player_id]
2565 self._print_sig_code(func, s)
2566 return func(s)
2567 except Exception as e:
2568 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2569
2570 def _decrypt_nsig(self, s, video_id, player_url):
2571 """Turn the encrypted n field into a working signature"""
2572 if player_url is None:
2573 raise ExtractorError('Cannot decrypt nsig without player_url')
2574 player_url = urljoin('https://www.youtube.com', player_url)
2575
2576 sig_id = ('nsig_value', s)
2577 if sig_id in self._player_cache:
2578 return self._player_cache[sig_id]
2579
2580 try:
2581 player_id = ('nsig', player_url)
2582 if player_id not in self._player_cache:
2583 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2584 func = self._player_cache[player_id]
2585 self._player_cache[sig_id] = func(s)
2586 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2587 return self._player_cache[sig_id]
2588 except Exception as e:
2589 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2590
2591 def _extract_n_function_name(self, jscode):
2592 nfunc, idx = self._search_regex(
2593 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2594 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2595 if not idx:
2596 return nfunc
2597 return json.loads(js_to_json(self._search_regex(
2598 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2599 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2600
2601 def _extract_n_function(self, video_id, player_url):
2602 player_id = self._extract_player_info(player_url)
2603 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2604
2605 if func_code:
2606 jsi = JSInterpreter(func_code)
2607 else:
2608 jscode = self._load_player(video_id, player_url)
2609 funcname = self._extract_n_function_name(jscode)
2610 jsi = JSInterpreter(jscode)
2611 func_code = jsi.extract_function_code(funcname)
2612 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2613
2614 if self.get_param('youtube_print_sig_code'):
2615 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2616
2617 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2618
2619 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2620 """
2621 Extract signatureTimestamp (sts)
2622 Required to tell API what sig/player version is in use.
2623 """
2624 sts = None
2625 if isinstance(ytcfg, dict):
2626 sts = int_or_none(ytcfg.get('STS'))
2627
2628 if not sts:
2629 # Attempt to extract from player
2630 if player_url is None:
2631 error_msg = 'Cannot extract signature timestamp without player_url.'
2632 if fatal:
2633 raise ExtractorError(error_msg)
2634 self.report_warning(error_msg)
2635 return
2636 code = self._load_player(video_id, player_url, fatal=fatal)
2637 if code:
2638 sts = int_or_none(self._search_regex(
2639 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2640 'JS player signature timestamp', group='sts', fatal=fatal))
2641 return sts
2642
2643 def _mark_watched(self, video_id, player_responses):
2644 playback_url = get_first(
2645 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2646 expected_type=url_or_none)
2647 if not playback_url:
2648 self.report_warning('Unable to mark watched')
2649 return
2650 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2651 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2652
2653 # cpn generation algorithm is reverse engineered from base.js.
2654 # In fact it works even with dummy cpn.
2655 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2656 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2657
2658 qs.update({
2659 'ver': ['2'],
2660 'cpn': [cpn],
2661 })
2662 playback_url = compat_urlparse.urlunparse(
2663 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2664
2665 self._download_webpage(
2666 playback_url, video_id, 'Marking watched',
2667 'Unable to mark watched', fatal=False)
2668
2669 @staticmethod
2670 def _extract_urls(webpage):
2671 # Embedded YouTube player
2672 entries = [
2673 unescapeHTML(mobj.group('url'))
2674 for mobj in re.finditer(r'''(?x)
2675 (?:
2676 <iframe[^>]+?src=|
2677 data-video-url=|
2678 <embed[^>]+?src=|
2679 embedSWF\(?:\s*|
2680 <object[^>]+data=|
2681 new\s+SWFObject\(
2682 )
2683 (["\'])
2684 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2685 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2686 \1''', webpage)]
2687
2688 # lazyYT YouTube embed
2689 entries.extend(list(map(
2690 unescapeHTML,
2691 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2692
2693 # Wordpress "YouTube Video Importer" plugin
2694 matches = re.findall(r'''(?x)<div[^>]+
2695 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2696 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2697 entries.extend(m[-1] for m in matches)
2698
2699 return entries
2700
2701 @staticmethod
2702 def _extract_url(webpage):
2703 urls = YoutubeIE._extract_urls(webpage)
2704 return urls[0] if urls else None
2705
2706 @classmethod
2707 def extract_id(cls, url):
2708 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2709 if mobj is None:
2710 raise ExtractorError('Invalid URL: %s' % url)
2711 return mobj.group('id')
2712
2713 def _extract_chapters_from_json(self, data, duration):
2714 chapter_list = traverse_obj(
2715 data, (
2716 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2717 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2718 ), expected_type=list)
2719
2720 return self._extract_chapters(
2721 chapter_list,
2722 chapter_time=lambda chapter: float_or_none(
2723 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2724 chapter_title=lambda chapter: traverse_obj(
2725 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2726 duration=duration)
2727
2728 def _extract_chapters_from_engagement_panel(self, data, duration):
2729 content_list = traverse_obj(
2730 data,
2731 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2732 expected_type=list, default=[])
2733 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2734 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2735
2736 return next(filter(None, (
2737 self._extract_chapters(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2738 chapter_time, chapter_title, duration)
2739 for contents in content_list)), [])
2740
2741 def _extract_chapters_from_description(self, description, duration):
2742 return self._extract_chapters(
2743 re.findall(r'(?m)^((?:\d+:)?\d{1,2}:\d{2})\b\W*\s(.+?)\s*$', description or ''),
2744 chapter_time=lambda x: parse_duration(x[0]), chapter_title=lambda x: x[1],
2745 duration=duration, strict=False)
2746
2747 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration, strict=True):
2748 if not duration:
2749 return
2750 chapter_list = [{
2751 'start_time': chapter_time(chapter),
2752 'title': chapter_title(chapter),
2753 } for chapter in chapter_list or []]
2754 if not strict:
2755 chapter_list.sort(key=lambda c: c['start_time'] or 0)
2756
2757 chapters = [{'start_time': 0, 'title': '<Untitled>'}]
2758 for idx, chapter in enumerate(chapter_list):
2759 if chapter['start_time'] is None or not chapter['title']:
2760 self.report_warning(f'Incomplete chapter {idx}')
2761 elif chapters[-1]['start_time'] <= chapter['start_time'] <= duration:
2762 chapters[-1]['end_time'] = chapter['start_time']
2763 chapters.append(chapter)
2764 else:
2765 self.report_warning(f'Invalid start time for chapter "{chapter["title"]}"')
2766 chapters[-1]['end_time'] = duration
2767 return chapters if len(chapters) > 1 and chapters[1]['start_time'] else chapters[1:]
2768
2769 def _extract_comment(self, comment_renderer, parent=None):
2770 comment_id = comment_renderer.get('commentId')
2771 if not comment_id:
2772 return
2773
2774 text = self._get_text(comment_renderer, 'contentText')
2775
2776 # note: timestamp is an estimate calculated from the current time and time_text
2777 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2778 author = self._get_text(comment_renderer, 'authorText')
2779 author_id = try_get(comment_renderer,
2780 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2781
2782 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2783 lambda x: x['likeCount']), compat_str)) or 0
2784 author_thumbnail = try_get(comment_renderer,
2785 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2786
2787 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2788 is_favorited = 'creatorHeart' in (try_get(
2789 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2790 return {
2791 'id': comment_id,
2792 'text': text,
2793 'timestamp': timestamp,
2794 'time_text': time_text,
2795 'like_count': votes,
2796 'is_favorited': is_favorited,
2797 'author': author,
2798 'author_id': author_id,
2799 'author_thumbnail': author_thumbnail,
2800 'author_is_uploader': author_is_uploader,
2801 'parent': parent or 'root'
2802 }
2803
2804 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2805
2806 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2807
2808 def extract_header(contents):
2809 _continuation = None
2810 for content in contents:
2811 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2812 expected_comment_count = self._get_count(
2813 comments_header_renderer, 'countText', 'commentsCount')
2814
2815 if expected_comment_count:
2816 tracker['est_total'] = expected_comment_count
2817 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2818 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2819
2820 sort_menu_item = try_get(
2821 comments_header_renderer,
2822 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2823 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2824
2825 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2826 if not _continuation:
2827 continue
2828
2829 sort_text = str_or_none(sort_menu_item.get('title'))
2830 if not sort_text:
2831 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2832 self.to_screen('Sorting comments by %s' % sort_text.lower())
2833 break
2834 return _continuation
2835
2836 def extract_thread(contents):
2837 if not parent:
2838 tracker['current_page_thread'] = 0
2839 for content in contents:
2840 if not parent and tracker['total_parent_comments'] >= max_parents:
2841 yield
2842 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2843 comment_renderer = get_first(
2844 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2845 expected_type=dict, default={})
2846
2847 comment = self._extract_comment(comment_renderer, parent)
2848 if not comment:
2849 continue
2850
2851 tracker['running_total'] += 1
2852 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2853 yield comment
2854
2855 # Attempt to get the replies
2856 comment_replies_renderer = try_get(
2857 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2858
2859 if comment_replies_renderer:
2860 tracker['current_page_thread'] += 1
2861 comment_entries_iter = self._comment_entries(
2862 comment_replies_renderer, ytcfg, video_id,
2863 parent=comment.get('id'), tracker=tracker)
2864 yield from itertools.islice(comment_entries_iter, min(
2865 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2866
2867 # Keeps track of counts across recursive calls
2868 if not tracker:
2869 tracker = dict(
2870 running_total=0,
2871 est_total=0,
2872 current_page_thread=0,
2873 total_parent_comments=0,
2874 total_reply_comments=0)
2875
2876 # TODO: Deprecated
2877 # YouTube comments have a max depth of 2
2878 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2879 if max_depth:
2880 self._downloader.deprecation_warning(
2881 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2882 if max_depth == 1 and parent:
2883 return
2884
2885 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2886 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2887
2888 continuation = self._extract_continuation(root_continuation_data)
2889
2890 response = None
2891 is_forced_continuation = False
2892 is_first_continuation = parent is None
2893 if is_first_continuation and not continuation:
2894 # Sometimes you can get comments by generating the continuation yourself,
2895 # even if YouTube initially reports them being disabled - e.g. stories comments.
2896 # Note: if the comment section is actually disabled, YouTube may return a response with
2897 # required check_get_keys missing. So we will disable that check initially in this case.
2898 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2899 is_forced_continuation = True
2900
2901 for page_num in itertools.count(0):
2902 if not continuation:
2903 break
2904 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2905 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2906 if page_num == 0:
2907 if is_first_continuation:
2908 note_prefix = 'Downloading comment section API JSON'
2909 else:
2910 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2911 tracker['current_page_thread'], comment_prog_str)
2912 else:
2913 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2914 ' ' if parent else '', ' replies' if parent else '',
2915 page_num, comment_prog_str)
2916
2917 response = self._extract_response(
2918 item_id=None, query=continuation,
2919 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2920 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2921 is_forced_continuation = False
2922 continuation_contents = traverse_obj(
2923 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2924
2925 continuation = None
2926 for continuation_section in continuation_contents:
2927 continuation_items = traverse_obj(
2928 continuation_section,
2929 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2930 get_all=False, expected_type=list) or []
2931 if is_first_continuation:
2932 continuation = extract_header(continuation_items)
2933 is_first_continuation = False
2934 if continuation:
2935 break
2936 continue
2937
2938 for entry in extract_thread(continuation_items):
2939 if not entry:
2940 return
2941 yield entry
2942 continuation = self._extract_continuation({'contents': continuation_items})
2943 if continuation:
2944 break
2945
2946 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2947 if message and not parent and tracker['running_total'] == 0:
2948 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2949
2950 @staticmethod
2951 def _generate_comment_continuation(video_id):
2952 """
2953 Generates initial comment section continuation token from given video id
2954 """
2955 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2956 return base64.b64encode(token.encode()).decode()
2957
2958 def _get_comments(self, ytcfg, video_id, contents, webpage):
2959 """Entry for comment extraction"""
2960 def _real_comment_extract(contents):
2961 renderer = next((
2962 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2963 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2964 yield from self._comment_entries(renderer, ytcfg, video_id)
2965
2966 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2967 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2968
2969 @staticmethod
2970 def _get_checkok_params():
2971 return {'contentCheckOk': True, 'racyCheckOk': True}
2972
2973 @classmethod
2974 def _generate_player_context(cls, sts=None):
2975 context = {
2976 'html5Preference': 'HTML5_PREF_WANTS',
2977 }
2978 if sts is not None:
2979 context['signatureTimestamp'] = sts
2980 return {
2981 'playbackContext': {
2982 'contentPlaybackContext': context
2983 },
2984 **cls._get_checkok_params()
2985 }
2986
2987 @staticmethod
2988 def _is_agegated(player_response):
2989 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2990 return True
2991
2992 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2993 AGE_GATE_REASONS = (
2994 'confirm your age', 'age-restricted', 'inappropriate', # reason
2995 'age_verification_required', 'age_check_required', # status
2996 )
2997 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2998
2999 @staticmethod
3000 def _is_unplayable(player_response):
3001 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3002
3003 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
3004
3005 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3006 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3007 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3008 headers = self.generate_api_headers(
3009 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3010
3011 yt_query = {
3012 'videoId': video_id,
3013 'params': '8AEB' # enable stories
3014 }
3015 yt_query.update(self._generate_player_context(sts))
3016 return self._extract_response(
3017 item_id=video_id, ep='player', query=yt_query,
3018 ytcfg=player_ytcfg, headers=headers, fatal=True,
3019 default_client=client,
3020 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3021 ) or None
3022
3023 def _get_requested_clients(self, url, smuggled_data):
3024 requested_clients = []
3025 default = ['android', 'web']
3026 allowed_clients = sorted(
3027 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3028 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3029 for client in self._configuration_arg('player_client'):
3030 if client in allowed_clients:
3031 requested_clients.append(client)
3032 elif client == 'default':
3033 requested_clients.extend(default)
3034 elif client == 'all':
3035 requested_clients.extend(allowed_clients)
3036 else:
3037 self.report_warning(f'Skipping unsupported client {client}')
3038 if not requested_clients:
3039 requested_clients = default
3040
3041 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3042 requested_clients.extend(
3043 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3044
3045 return orderedSet(requested_clients)
3046
3047 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3048 initial_pr = None
3049 if webpage:
3050 initial_pr = self._search_json(
3051 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3052
3053 all_clients = set(clients)
3054 clients = clients[::-1]
3055 prs = []
3056
3057 def append_client(*client_names):
3058 """ Append the first client name that exists but not already used """
3059 for client_name in client_names:
3060 actual_client = _split_innertube_client(client_name)[0]
3061 if actual_client in INNERTUBE_CLIENTS:
3062 if actual_client not in all_clients:
3063 clients.append(client_name)
3064 all_clients.add(actual_client)
3065 return
3066
3067 # Android player_response does not have microFormats which are needed for
3068 # extraction of some data. So we return the initial_pr with formats
3069 # stripped out even if not requested by the user
3070 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3071 if initial_pr:
3072 pr = dict(initial_pr)
3073 pr['streamingData'] = None
3074 prs.append(pr)
3075
3076 last_error = None
3077 tried_iframe_fallback = False
3078 player_url = None
3079 while clients:
3080 client, base_client, variant = _split_innertube_client(clients.pop())
3081 player_ytcfg = master_ytcfg if client == 'web' else {}
3082 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3083 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3084
3085 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3086 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3087 if 'js' in self._configuration_arg('player_skip'):
3088 require_js_player = False
3089 player_url = None
3090
3091 if not player_url and not tried_iframe_fallback and require_js_player:
3092 player_url = self._download_player_url(video_id)
3093 tried_iframe_fallback = True
3094
3095 try:
3096 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3097 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3098 except ExtractorError as e:
3099 if last_error:
3100 self.report_warning(last_error)
3101 last_error = e
3102 continue
3103
3104 if pr:
3105 prs.append(pr)
3106
3107 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3108 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3109 append_client(f'{base_client}_creator')
3110 elif self._is_agegated(pr):
3111 if variant == 'tv_embedded':
3112 append_client(f'{base_client}_embedded')
3113 elif not variant:
3114 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3115
3116 if last_error:
3117 if not len(prs):
3118 raise last_error
3119 self.report_warning(last_error)
3120 return prs, player_url
3121
3122 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3123 itags, stream_ids = {}, []
3124 itag_qualities, res_qualities = {}, {}
3125 q = qualities([
3126 # Normally tiny is the smallest video-only formats. But
3127 # audio-only formats with unknown quality may get tagged as tiny
3128 'tiny',
3129 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3130 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3131 ])
3132 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3133
3134 for fmt in streaming_formats:
3135 if fmt.get('targetDurationSec'):
3136 continue
3137
3138 itag = str_or_none(fmt.get('itag'))
3139 audio_track = fmt.get('audioTrack') or {}
3140 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3141 if stream_id in stream_ids:
3142 continue
3143
3144 quality = fmt.get('quality')
3145 height = int_or_none(fmt.get('height'))
3146 if quality == 'tiny' or not quality:
3147 quality = fmt.get('audioQuality', '').lower() or quality
3148 # The 3gp format (17) in android client has a quality of "small",
3149 # but is actually worse than other formats
3150 if itag == '17':
3151 quality = 'tiny'
3152 if quality:
3153 if itag:
3154 itag_qualities[itag] = quality
3155 if height:
3156 res_qualities[height] = quality
3157 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3158 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3159 # number of fragment that would subsequently requested with (`&sq=N`)
3160 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3161 continue
3162
3163 fmt_url = fmt.get('url')
3164 if not fmt_url:
3165 sc = compat_parse_qs(fmt.get('signatureCipher'))
3166 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3167 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3168 if not all((sc, fmt_url, player_url, encrypted_sig)):
3169 continue
3170 try:
3171 fmt_url += '&%s=%s' % (
3172 traverse_obj(sc, ('sp', -1)) or 'signature',
3173 self._decrypt_signature(encrypted_sig, video_id, player_url)
3174 )
3175 except ExtractorError as e:
3176 self.report_warning('Signature extraction failed: Some formats may be missing', only_once=True)
3177 self.write_debug(e, only_once=True)
3178 continue
3179
3180 query = parse_qs(fmt_url)
3181 throttled = False
3182 if query.get('n'):
3183 try:
3184 fmt_url = update_url_query(fmt_url, {
3185 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3186 except ExtractorError as e:
3187 self.report_warning(
3188 'nsig extraction failed: You may experience throttling for some formats\n'
3189 f'n = {query["n"][0]} ; player = {player_url}', only_once=True)
3190 self.write_debug(e, only_once=True)
3191 throttled = True
3192
3193 if itag:
3194 itags[itag] = 'https'
3195 stream_ids.append(stream_id)
3196
3197 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3198 language_preference = (
3199 10 if audio_track.get('audioIsDefault') and 10
3200 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3201 else -1)
3202 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3203 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3204 # Make sure to avoid false positives with small duration differences.
3205 # Eg: __2ABJjxzNo, ySuUZEjARPY
3206 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3207 if is_damaged:
3208 self.report_warning(
3209 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3210 dct = {
3211 'asr': int_or_none(fmt.get('audioSampleRate')),
3212 'filesize': int_or_none(fmt.get('contentLength')),
3213 'format_id': itag,
3214 'format_note': join_nonempty(
3215 '%s%s' % (audio_track.get('displayName') or '',
3216 ' (default)' if language_preference > 0 else ''),
3217 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3218 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3219 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3220 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3221 'fps': int_or_none(fmt.get('fps')) or None,
3222 'height': height,
3223 'quality': q(quality),
3224 'has_drm': bool(fmt.get('drmFamilies')),
3225 'tbr': tbr,
3226 'url': fmt_url,
3227 'width': int_or_none(fmt.get('width')),
3228 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3229 'desc' if language_preference < -1 else ''),
3230 'language_preference': language_preference,
3231 # Strictly de-prioritize damaged and 3gp formats
3232 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3233 }
3234 mime_mobj = re.match(
3235 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3236 if mime_mobj:
3237 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3238 dct.update(parse_codecs(mime_mobj.group(2)))
3239 no_audio = dct.get('acodec') == 'none'
3240 no_video = dct.get('vcodec') == 'none'
3241 if no_audio:
3242 dct['vbr'] = tbr
3243 if no_video:
3244 dct['abr'] = tbr
3245 if no_audio or no_video:
3246 dct['downloader_options'] = {
3247 # Youtube throttles chunks >~10M
3248 'http_chunk_size': 10485760,
3249 }
3250 if dct.get('ext'):
3251 dct['container'] = dct['ext'] + '_dash'
3252 yield dct
3253
3254 live_from_start = is_live and self.get_param('live_from_start')
3255 skip_manifests = self._configuration_arg('skip')
3256 if not self.get_param('youtube_include_hls_manifest', True):
3257 skip_manifests.append('hls')
3258 if not self.get_param('youtube_include_dash_manifest', True):
3259 skip_manifests.append('dash')
3260 get_dash = 'dash' not in skip_manifests and (
3261 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3262 get_hls = not live_from_start and 'hls' not in skip_manifests
3263
3264 def process_manifest_format(f, proto, itag):
3265 if itag in itags:
3266 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3267 return False
3268 itag = f'{itag}-{proto}'
3269 if itag:
3270 f['format_id'] = itag
3271 itags[itag] = proto
3272
3273 f['quality'] = next((
3274 q(qdict[val])
3275 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3276 if val in qdict), -1)
3277 return True
3278
3279 for sd in streaming_data:
3280 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3281 if hls_manifest_url:
3282 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3283 if process_manifest_format(f, 'hls', self._search_regex(
3284 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3285 yield f
3286
3287 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3288 if dash_manifest_url:
3289 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3290 if process_manifest_format(f, 'dash', f['format_id']):
3291 f['filesize'] = int_or_none(self._search_regex(
3292 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3293 if live_from_start:
3294 f['is_from_start'] = True
3295
3296 yield f
3297
3298 def _extract_storyboard(self, player_responses, duration):
3299 spec = get_first(
3300 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3301 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3302 if not base_url:
3303 return
3304 L = len(spec) - 1
3305 for i, args in enumerate(spec):
3306 args = args.split('#')
3307 counts = list(map(int_or_none, args[:5]))
3308 if len(args) != 8 or not all(counts):
3309 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3310 continue
3311 width, height, frame_count, cols, rows = counts
3312 N, sigh = args[6:]
3313
3314 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3315 fragment_count = frame_count / (cols * rows)
3316 fragment_duration = duration / fragment_count
3317 yield {
3318 'format_id': f'sb{i}',
3319 'format_note': 'storyboard',
3320 'ext': 'mhtml',
3321 'protocol': 'mhtml',
3322 'acodec': 'none',
3323 'vcodec': 'none',
3324 'url': url,
3325 'width': width,
3326 'height': height,
3327 'fragments': [{
3328 'url': url.replace('$M', str(j)),
3329 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3330 } for j in range(math.ceil(fragment_count))],
3331 }
3332
3333 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3334 webpage = None
3335 if 'webpage' not in self._configuration_arg('player_skip'):
3336 webpage = self._download_webpage(
3337 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3338
3339 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3340
3341 player_responses, player_url = self._extract_player_responses(
3342 self._get_requested_clients(url, smuggled_data),
3343 video_id, webpage, master_ytcfg)
3344
3345 return webpage, master_ytcfg, player_responses, player_url
3346
3347 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3348 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3349 is_live = get_first(video_details, 'isLive')
3350 if is_live is None:
3351 is_live = get_first(live_broadcast_details, 'isLiveNow')
3352
3353 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3354 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3355
3356 return live_broadcast_details, is_live, streaming_data, formats
3357
3358 def _real_extract(self, url):
3359 url, smuggled_data = unsmuggle_url(url, {})
3360 video_id = self._match_id(url)
3361
3362 base_url = self.http_scheme() + '//www.youtube.com/'
3363 webpage_url = base_url + 'watch?v=' + video_id
3364
3365 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3366
3367 playability_statuses = traverse_obj(
3368 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3369
3370 trailer_video_id = get_first(
3371 playability_statuses,
3372 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3373 expected_type=str)
3374 if trailer_video_id:
3375 return self.url_result(
3376 trailer_video_id, self.ie_key(), trailer_video_id)
3377
3378 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3379 if webpage else (lambda x: None))
3380
3381 video_details = traverse_obj(
3382 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3383 microformats = traverse_obj(
3384 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3385 expected_type=dict, default=[])
3386 video_title = (
3387 get_first(video_details, 'title')
3388 or self._get_text(microformats, (..., 'title'))
3389 or search_meta(['og:title', 'twitter:title', 'title']))
3390 video_description = get_first(video_details, 'shortDescription')
3391
3392 multifeed_metadata_list = get_first(
3393 player_responses,
3394 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3395 expected_type=str)
3396 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3397 if self.get_param('noplaylist'):
3398 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3399 else:
3400 entries = []
3401 feed_ids = []
3402 for feed in multifeed_metadata_list.split(','):
3403 # Unquote should take place before split on comma (,) since textual
3404 # fields may contain comma as well (see
3405 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3406 feed_data = compat_parse_qs(
3407 compat_urllib_parse_unquote_plus(feed))
3408
3409 def feed_entry(name):
3410 return try_get(
3411 feed_data, lambda x: x[name][0], compat_str)
3412
3413 feed_id = feed_entry('id')
3414 if not feed_id:
3415 continue
3416 feed_title = feed_entry('title')
3417 title = video_title
3418 if feed_title:
3419 title += ' (%s)' % feed_title
3420 entries.append({
3421 '_type': 'url_transparent',
3422 'ie_key': 'Youtube',
3423 'url': smuggle_url(
3424 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3425 {'force_singlefeed': True}),
3426 'title': title,
3427 })
3428 feed_ids.append(feed_id)
3429 self.to_screen(
3430 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3431 % (', '.join(feed_ids), video_id))
3432 return self.playlist_result(
3433 entries, video_id, video_title, video_description)
3434
3435 duration = int_or_none(
3436 get_first(video_details, 'lengthSeconds')
3437 or get_first(microformats, 'lengthSeconds')
3438 or parse_duration(search_meta('duration'))) or None
3439
3440 if get_first(video_details, 'isPostLiveDvr'):
3441 self.write_debug('Video is in Post-Live Manifestless mode')
3442 if duration or 0 > 4 * 3600:
3443 self.report_warning(
3444 'The livestream has not finished processing. Only 4 hours of the video can be currently downloaded. '
3445 'This is a known issue and patches are welcome')
3446
3447 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3448 video_id, microformats, video_details, player_responses, player_url, duration)
3449
3450 if not formats:
3451 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3452 self.report_drm(video_id)
3453 pemr = get_first(
3454 playability_statuses,
3455 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3456 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3457 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3458 if subreason:
3459 if subreason == 'The uploader has not made this video available in your country.':
3460 countries = get_first(microformats, 'availableCountries')
3461 if not countries:
3462 regions_allowed = search_meta('regionsAllowed')
3463 countries = regions_allowed.split(',') if regions_allowed else None
3464 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3465 reason += f'. {subreason}'
3466 if reason:
3467 self.raise_no_formats(reason, expected=True)
3468
3469 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3470 if not keywords and webpage:
3471 keywords = [
3472 unescapeHTML(m.group('content'))
3473 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3474 for keyword in keywords:
3475 if keyword.startswith('yt:stretch='):
3476 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3477 if mobj:
3478 # NB: float is intentional for forcing float division
3479 w, h = (float(v) for v in mobj.groups())
3480 if w > 0 and h > 0:
3481 ratio = w / h
3482 for f in formats:
3483 if f.get('vcodec') != 'none':
3484 f['stretched_ratio'] = ratio
3485 break
3486 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3487 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3488 if thumbnail_url:
3489 thumbnails.append({
3490 'url': thumbnail_url,
3491 })
3492 original_thumbnails = thumbnails.copy()
3493
3494 # The best resolution thumbnails sometimes does not appear in the webpage
3495 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3496 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3497 thumbnail_names = [
3498 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3499 # in resolution, these are not the custom thumbnail. So de-prioritize them
3500 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3501 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3502 ]
3503 n_thumbnail_names = len(thumbnail_names)
3504 thumbnails.extend({
3505 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3506 video_id=video_id, name=name, ext=ext,
3507 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3508 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3509 for thumb in thumbnails:
3510 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3511 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3512 self._remove_duplicate_formats(thumbnails)
3513 self._downloader._sort_thumbnails(original_thumbnails)
3514
3515 category = get_first(microformats, 'category') or search_meta('genre')
3516 channel_id = str_or_none(
3517 get_first(video_details, 'channelId')
3518 or get_first(microformats, 'externalChannelId')
3519 or search_meta('channelId'))
3520 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3521
3522 live_content = get_first(video_details, 'isLiveContent')
3523 is_upcoming = get_first(video_details, 'isUpcoming')
3524 if is_live is None:
3525 if is_upcoming or live_content is False:
3526 is_live = False
3527 if is_upcoming is None and (live_content or is_live):
3528 is_upcoming = False
3529 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3530 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3531 if not duration and live_end_time and live_start_time:
3532 duration = live_end_time - live_start_time
3533
3534 if is_live and self.get_param('live_from_start'):
3535 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3536
3537 formats.extend(self._extract_storyboard(player_responses, duration))
3538
3539 # Source is given priority since formats that throttle are given lower source_preference
3540 # When throttling issue is fully fixed, remove this
3541 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3542
3543 info = {
3544 'id': video_id,
3545 'title': video_title,
3546 'formats': formats,
3547 'thumbnails': thumbnails,
3548 # The best thumbnail that we are sure exists. Prevents unnecessary
3549 # URL checking if user don't care about getting the best possible thumbnail
3550 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3551 'description': video_description,
3552 'uploader': get_first(video_details, 'author'),
3553 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3554 'uploader_url': owner_profile_url,
3555 'channel_id': channel_id,
3556 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3557 'duration': duration,
3558 'view_count': int_or_none(
3559 get_first((video_details, microformats), (..., 'viewCount'))
3560 or search_meta('interactionCount')),
3561 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3562 'age_limit': 18 if (
3563 get_first(microformats, 'isFamilySafe') is False
3564 or search_meta('isFamilyFriendly') == 'false'
3565 or search_meta('og:restrictions:age') == '18+') else 0,
3566 'webpage_url': webpage_url,
3567 'categories': [category] if category else None,
3568 'tags': keywords,
3569 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3570 'is_live': is_live,
3571 'was_live': (False if is_live or is_upcoming or live_content is False
3572 else None if is_live is None or is_upcoming is None
3573 else live_content),
3574 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3575 'release_timestamp': live_start_time,
3576 }
3577
3578 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3579 if pctr:
3580 def get_lang_code(track):
3581 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3582 or track.get('languageCode'))
3583
3584 # Converted into dicts to remove duplicates
3585 captions = {
3586 get_lang_code(sub): sub
3587 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3588 translation_languages = {
3589 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3590 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3591
3592 def process_language(container, base_url, lang_code, sub_name, query):
3593 lang_subs = container.setdefault(lang_code, [])
3594 for fmt in self._SUBTITLE_FORMATS:
3595 query.update({
3596 'fmt': fmt,
3597 })
3598 lang_subs.append({
3599 'ext': fmt,
3600 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3601 'name': sub_name,
3602 })
3603
3604 subtitles, automatic_captions = {}, {}
3605 for lang_code, caption_track in captions.items():
3606 base_url = caption_track.get('baseUrl')
3607 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3608 if not base_url:
3609 continue
3610 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3611 if caption_track.get('kind') != 'asr':
3612 if not lang_code:
3613 continue
3614 process_language(
3615 subtitles, base_url, lang_code, lang_name, {})
3616 if not caption_track.get('isTranslatable'):
3617 continue
3618 for trans_code, trans_name in translation_languages.items():
3619 if not trans_code:
3620 continue
3621 orig_trans_code = trans_code
3622 if caption_track.get('kind') != 'asr':
3623 if 'translated_subs' in self._configuration_arg('skip'):
3624 continue
3625 trans_code += f'-{lang_code}'
3626 trans_name += format_field(lang_name, template=' from %s')
3627 # Add an "-orig" label to the original language so that it can be distinguished.
3628 # The subs are returned without "-orig" as well for compatibility
3629 if lang_code == f'a-{orig_trans_code}':
3630 process_language(
3631 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3632 # Setting tlang=lang returns damaged subtitles.
3633 process_language(automatic_captions, base_url, trans_code, trans_name,
3634 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3635 info['automatic_captions'] = automatic_captions
3636 info['subtitles'] = subtitles
3637
3638 parsed_url = compat_urllib_parse_urlparse(url)
3639 for component in [parsed_url.fragment, parsed_url.query]:
3640 query = compat_parse_qs(component)
3641 for k, v in query.items():
3642 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3643 d_k += '_time'
3644 if d_k not in info and k in s_ks:
3645 info[d_k] = parse_duration(query[k][0])
3646
3647 # Youtube Music Auto-generated description
3648 if video_description:
3649 mobj = re.search(
3650 r'''(?xs)
3651 (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
3652 (?P<album>[^\n]+)
3653 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
3654 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
3655 (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
3656 .+\nAuto-generated\ by\ YouTube\.\s*$
3657 ''', video_description)
3658 if mobj:
3659 release_year = mobj.group('release_year')
3660 release_date = mobj.group('release_date')
3661 if release_date:
3662 release_date = release_date.replace('-', '')
3663 if not release_year:
3664 release_year = release_date[:4]
3665 info.update({
3666 'album': mobj.group('album'.strip()),
3667 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3668 'track': mobj.group('track').strip(),
3669 'release_date': release_date,
3670 'release_year': int_or_none(release_year),
3671 })
3672
3673 initial_data = None
3674 if webpage:
3675 initial_data = self._search_json(
3676 self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', video_id, fatal=False)
3677 if not initial_data:
3678 query = {'videoId': video_id}
3679 query.update(self._get_checkok_params())
3680 initial_data = self._extract_response(
3681 item_id=video_id, ep='next', fatal=False,
3682 ytcfg=master_ytcfg, query=query,
3683 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3684 note='Downloading initial data API JSON')
3685
3686 try: # This will error if there is no livechat
3687 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3688 except (KeyError, IndexError, TypeError):
3689 pass
3690 else:
3691 info.setdefault('subtitles', {})['live_chat'] = [{
3692 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3693 'video_id': video_id,
3694 'ext': 'json',
3695 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3696 }]
3697
3698 if initial_data:
3699 info['chapters'] = (
3700 self._extract_chapters_from_json(initial_data, duration)
3701 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3702 or self._extract_chapters_from_description(video_description, duration)
3703 or None)
3704
3705 contents = traverse_obj(
3706 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3707 expected_type=list, default=[])
3708
3709 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3710 if vpir:
3711 stl = vpir.get('superTitleLink')
3712 if stl:
3713 stl = self._get_text(stl)
3714 if try_get(
3715 vpir,
3716 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3717 info['location'] = stl
3718 else:
3719 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3720 if mobj:
3721 info.update({
3722 'series': mobj.group(1),
3723 'season_number': int(mobj.group(2)),
3724 'episode_number': int(mobj.group(3)),
3725 })
3726 for tlb in (try_get(
3727 vpir,
3728 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3729 list) or []):
3730 tbr = tlb.get('toggleButtonRenderer') or {}
3731 for getter, regex in [(
3732 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3733 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3734 lambda x: x['accessibility'],
3735 lambda x: x['accessibilityData']['accessibilityData'],
3736 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3737 label = (try_get(tbr, getter, dict) or {}).get('label')
3738 if label:
3739 mobj = re.match(regex, label)
3740 if mobj:
3741 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3742 break
3743 sbr_tooltip = try_get(
3744 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3745 if sbr_tooltip:
3746 like_count, dislike_count = sbr_tooltip.split(' / ')
3747 info.update({
3748 'like_count': str_to_int(like_count),
3749 'dislike_count': str_to_int(dislike_count),
3750 })
3751 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3752 if vsir:
3753 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3754 info.update({
3755 'channel': self._get_text(vor, 'title'),
3756 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3757
3758 rows = try_get(
3759 vsir,
3760 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3761 list) or []
3762 multiple_songs = False
3763 for row in rows:
3764 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3765 multiple_songs = True
3766 break
3767 for row in rows:
3768 mrr = row.get('metadataRowRenderer') or {}
3769 mrr_title = mrr.get('title')
3770 if not mrr_title:
3771 continue
3772 mrr_title = self._get_text(mrr, 'title')
3773 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3774 if mrr_title == 'License':
3775 info['license'] = mrr_contents_text
3776 elif not multiple_songs:
3777 if mrr_title == 'Album':
3778 info['album'] = mrr_contents_text
3779 elif mrr_title == 'Artist':
3780 info['artist'] = mrr_contents_text
3781 elif mrr_title == 'Song':
3782 info['track'] = mrr_contents_text
3783
3784 fallbacks = {
3785 'channel': 'uploader',
3786 'channel_id': 'uploader_id',
3787 'channel_url': 'uploader_url',
3788 }
3789
3790 # The upload date for scheduled, live and past live streams / premieres in microformats
3791 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3792 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3793 upload_date = (
3794 unified_strdate(get_first(microformats, 'uploadDate'))
3795 or unified_strdate(search_meta('uploadDate')))
3796 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3797 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3798 info['upload_date'] = upload_date
3799
3800 for to, frm in fallbacks.items():
3801 if not info.get(to):
3802 info[to] = info.get(frm)
3803
3804 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3805 v = info.get(s_k)
3806 if v:
3807 info[d_k] = v
3808
3809 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3810 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3811 is_membersonly = None
3812 is_premium = None
3813 if initial_data and is_private is not None:
3814 is_membersonly = False
3815 is_premium = False
3816 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3817 badge_labels = set()
3818 for content in contents:
3819 if not isinstance(content, dict):
3820 continue
3821 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3822 for badge_label in badge_labels:
3823 if badge_label.lower() == 'members only':
3824 is_membersonly = True
3825 elif badge_label.lower() == 'premium':
3826 is_premium = True
3827 elif badge_label.lower() == 'unlisted':
3828 is_unlisted = True
3829
3830 info['availability'] = self._availability(
3831 is_private=is_private,
3832 needs_premium=is_premium,
3833 needs_subscription=is_membersonly,
3834 needs_auth=info['age_limit'] >= 18,
3835 is_unlisted=None if is_private is None else is_unlisted)
3836
3837 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3838
3839 self.mark_watched(video_id, player_responses)
3840
3841 return info
3842
3843
3844class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3845
3846 @staticmethod
3847 def passthrough_smuggled_data(func):
3848 def _smuggle(entries, smuggled_data):
3849 for entry in entries:
3850 # TODO: Convert URL to music.youtube instead.
3851 # Do we need to passthrough any other smuggled_data?
3852 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3853 yield entry
3854
3855 @functools.wraps(func)
3856 def wrapper(self, url):
3857 url, smuggled_data = unsmuggle_url(url, {})
3858 if self.is_music_url(url):
3859 smuggled_data['is_music_url'] = True
3860 info_dict = func(self, url, smuggled_data)
3861 if smuggled_data and info_dict.get('entries'):
3862 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3863 return info_dict
3864 return wrapper
3865
3866 def _extract_channel_id(self, webpage):
3867 channel_id = self._html_search_meta(
3868 'channelId', webpage, 'channel id', default=None)
3869 if channel_id:
3870 return channel_id
3871 channel_url = self._html_search_meta(
3872 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3873 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3874 'twitter:app:url:googleplay'), webpage, 'channel url')
3875 return self._search_regex(
3876 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3877 channel_url, 'channel id')
3878
3879 @staticmethod
3880 def _extract_basic_item_renderer(item):
3881 # Modified from _extract_grid_item_renderer
3882 known_basic_renderers = (
3883 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3884 )
3885 for key, renderer in item.items():
3886 if not isinstance(renderer, dict):
3887 continue
3888 elif key in known_basic_renderers:
3889 return renderer
3890 elif key.startswith('grid') and key.endswith('Renderer'):
3891 return renderer
3892
3893 def _grid_entries(self, grid_renderer):
3894 for item in grid_renderer['items']:
3895 if not isinstance(item, dict):
3896 continue
3897 renderer = self._extract_basic_item_renderer(item)
3898 if not isinstance(renderer, dict):
3899 continue
3900 title = self._get_text(renderer, 'title')
3901
3902 # playlist
3903 playlist_id = renderer.get('playlistId')
3904 if playlist_id:
3905 yield self.url_result(
3906 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3907 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3908 video_title=title)
3909 continue
3910 # video
3911 video_id = renderer.get('videoId')
3912 if video_id:
3913 yield self._extract_video(renderer)
3914 continue
3915 # channel
3916 channel_id = renderer.get('channelId')
3917 if channel_id:
3918 yield self.url_result(
3919 'https://www.youtube.com/channel/%s' % channel_id,
3920 ie=YoutubeTabIE.ie_key(), video_title=title)
3921 continue
3922 # generic endpoint URL support
3923 ep_url = urljoin('https://www.youtube.com/', try_get(
3924 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3925 compat_str))
3926 if ep_url:
3927 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3928 if ie.suitable(ep_url):
3929 yield self.url_result(
3930 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3931 break
3932
3933 def _music_reponsive_list_entry(self, renderer):
3934 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3935 if video_id:
3936 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3937 ie=YoutubeIE.ie_key(), video_id=video_id)
3938 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3939 if playlist_id:
3940 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3941 if video_id:
3942 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3943 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3944 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3945 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3946 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3947 if browse_id:
3948 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3949 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3950
3951 def _shelf_entries_from_content(self, shelf_renderer):
3952 content = shelf_renderer.get('content')
3953 if not isinstance(content, dict):
3954 return
3955 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3956 if renderer:
3957 # TODO: add support for nested playlists so each shelf is processed
3958 # as separate playlist
3959 # TODO: this includes only first N items
3960 yield from self._grid_entries(renderer)
3961 renderer = content.get('horizontalListRenderer')
3962 if renderer:
3963 # TODO
3964 pass
3965
3966 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3967 ep = try_get(
3968 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3969 compat_str)
3970 shelf_url = urljoin('https://www.youtube.com', ep)
3971 if shelf_url:
3972 # Skipping links to another channels, note that checking for
3973 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3974 # will not work
3975 if skip_channels and '/channels?' in shelf_url:
3976 return
3977 title = self._get_text(shelf_renderer, 'title')
3978 yield self.url_result(shelf_url, video_title=title)
3979 # Shelf may not contain shelf URL, fallback to extraction from content
3980 yield from self._shelf_entries_from_content(shelf_renderer)
3981
3982 def _playlist_entries(self, video_list_renderer):
3983 for content in video_list_renderer['contents']:
3984 if not isinstance(content, dict):
3985 continue
3986 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3987 if not isinstance(renderer, dict):
3988 continue
3989 video_id = renderer.get('videoId')
3990 if not video_id:
3991 continue
3992 yield self._extract_video(renderer)
3993
3994 def _rich_entries(self, rich_grid_renderer):
3995 renderer = try_get(
3996 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3997 video_id = renderer.get('videoId')
3998 if not video_id:
3999 return
4000 yield self._extract_video(renderer)
4001
4002 def _video_entry(self, video_renderer):
4003 video_id = video_renderer.get('videoId')
4004 if video_id:
4005 return self._extract_video(video_renderer)
4006
4007 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4008 url = urljoin('https://youtube.com', traverse_obj(
4009 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4010 if url:
4011 return self.url_result(
4012 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4013
4014 def _post_thread_entries(self, post_thread_renderer):
4015 post_renderer = try_get(
4016 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4017 if not post_renderer:
4018 return
4019 # video attachment
4020 video_renderer = try_get(
4021 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4022 video_id = video_renderer.get('videoId')
4023 if video_id:
4024 entry = self._extract_video(video_renderer)
4025 if entry:
4026 yield entry
4027 # playlist attachment
4028 playlist_id = try_get(
4029 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
4030 if playlist_id:
4031 yield self.url_result(
4032 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4033 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4034 # inline video links
4035 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4036 for run in runs:
4037 if not isinstance(run, dict):
4038 continue
4039 ep_url = try_get(
4040 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
4041 if not ep_url:
4042 continue
4043 if not YoutubeIE.suitable(ep_url):
4044 continue
4045 ep_video_id = YoutubeIE._match_id(ep_url)
4046 if video_id == ep_video_id:
4047 continue
4048 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4049
4050 def _post_thread_continuation_entries(self, post_thread_continuation):
4051 contents = post_thread_continuation.get('contents')
4052 if not isinstance(contents, list):
4053 return
4054 for content in contents:
4055 renderer = content.get('backstagePostThreadRenderer')
4056 if isinstance(renderer, dict):
4057 yield from self._post_thread_entries(renderer)
4058 continue
4059 renderer = content.get('videoRenderer')
4060 if isinstance(renderer, dict):
4061 yield self._video_entry(renderer)
4062
4063 r''' # unused
4064 def _rich_grid_entries(self, contents):
4065 for content in contents:
4066 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4067 if video_renderer:
4068 entry = self._video_entry(video_renderer)
4069 if entry:
4070 yield entry
4071 '''
4072
4073 def _extract_entries(self, parent_renderer, continuation_list):
4074 # continuation_list is modified in-place with continuation_list = [continuation_token]
4075 continuation_list[:] = [None]
4076 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4077 for content in contents:
4078 if not isinstance(content, dict):
4079 continue
4080 is_renderer = traverse_obj(
4081 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4082 expected_type=dict)
4083 if not is_renderer:
4084 renderer = content.get('richItemRenderer')
4085 if renderer:
4086 for entry in self._rich_entries(renderer):
4087 yield entry
4088 continuation_list[0] = self._extract_continuation(parent_renderer)
4089 continue
4090 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4091 for isr_content in isr_contents:
4092 if not isinstance(isr_content, dict):
4093 continue
4094
4095 known_renderers = {
4096 'playlistVideoListRenderer': self._playlist_entries,
4097 'gridRenderer': self._grid_entries,
4098 'reelShelfRenderer': self._grid_entries,
4099 'shelfRenderer': self._shelf_entries,
4100 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4101 'backstagePostThreadRenderer': self._post_thread_entries,
4102 'videoRenderer': lambda x: [self._video_entry(x)],
4103 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4104 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4105 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4106 }
4107 for key, renderer in isr_content.items():
4108 if key not in known_renderers:
4109 continue
4110 for entry in known_renderers[key](renderer):
4111 if entry:
4112 yield entry
4113 continuation_list[0] = self._extract_continuation(renderer)
4114 break
4115
4116 if not continuation_list[0]:
4117 continuation_list[0] = self._extract_continuation(is_renderer)
4118
4119 if not continuation_list[0]:
4120 continuation_list[0] = self._extract_continuation(parent_renderer)
4121
4122 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4123 continuation_list = [None]
4124 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4125 tab_content = try_get(tab, lambda x: x['content'], dict)
4126 if not tab_content:
4127 return
4128 parent_renderer = (
4129 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4130 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4131 yield from extract_entries(parent_renderer)
4132 continuation = continuation_list[0]
4133
4134 for page_num in itertools.count(1):
4135 if not continuation:
4136 break
4137 headers = self.generate_api_headers(
4138 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4139 response = self._extract_response(
4140 item_id=f'{item_id} page {page_num}',
4141 query=continuation, headers=headers, ytcfg=ytcfg,
4142 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4143
4144 if not response:
4145 break
4146 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4147 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4148 visitor_data = self._extract_visitor_data(response) or visitor_data
4149
4150 known_continuation_renderers = {
4151 'playlistVideoListContinuation': self._playlist_entries,
4152 'gridContinuation': self._grid_entries,
4153 'itemSectionContinuation': self._post_thread_continuation_entries,
4154 'sectionListContinuation': extract_entries, # for feeds
4155 }
4156 continuation_contents = try_get(
4157 response, lambda x: x['continuationContents'], dict) or {}
4158 continuation_renderer = None
4159 for key, value in continuation_contents.items():
4160 if key not in known_continuation_renderers:
4161 continue
4162 continuation_renderer = value
4163 continuation_list = [None]
4164 yield from known_continuation_renderers[key](continuation_renderer)
4165 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4166 break
4167 if continuation_renderer:
4168 continue
4169
4170 known_renderers = {
4171 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4172 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4173 'gridVideoRenderer': (self._grid_entries, 'items'),
4174 'gridChannelRenderer': (self._grid_entries, 'items'),
4175 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4176 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4177 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4178 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4179 }
4180 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4181 continuation_items = try_get(
4182 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4183 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4184 video_items_renderer = None
4185 for key, value in continuation_item.items():
4186 if key not in known_renderers:
4187 continue
4188 video_items_renderer = {known_renderers[key][1]: continuation_items}
4189 continuation_list = [None]
4190 yield from known_renderers[key][0](video_items_renderer)
4191 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4192 break
4193 if video_items_renderer:
4194 continue
4195 break
4196
4197 @staticmethod
4198 def _extract_selected_tab(tabs, fatal=True):
4199 for tab in tabs:
4200 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4201 if renderer.get('selected') is True:
4202 return renderer
4203 else:
4204 if fatal:
4205 raise ExtractorError('Unable to find selected tab')
4206
4207 def _extract_uploader(self, data):
4208 uploader = {}
4209 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4210 owner = try_get(
4211 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4212 if owner:
4213 owner_text = owner.get('text')
4214 uploader['uploader'] = self._search_regex(
4215 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4216 uploader['uploader_id'] = try_get(
4217 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4218 uploader['uploader_url'] = urljoin(
4219 'https://www.youtube.com/',
4220 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4221 return {k: v for k, v in uploader.items() if v is not None}
4222
4223 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4224 playlist_id = title = description = channel_url = channel_name = channel_id = None
4225 tags = []
4226
4227 selected_tab = self._extract_selected_tab(tabs)
4228 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4229 renderer = try_get(
4230 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4231 if renderer:
4232 channel_name = renderer.get('title')
4233 channel_url = renderer.get('channelUrl')
4234 channel_id = renderer.get('externalId')
4235 else:
4236 renderer = try_get(
4237 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4238
4239 if renderer:
4240 title = renderer.get('title')
4241 description = renderer.get('description', '')
4242 playlist_id = channel_id
4243 tags = renderer.get('keywords', '').split()
4244
4245 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4246 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4247 def _get_uncropped(url):
4248 return url_or_none((url or '').split('=')[0] + '=s0')
4249
4250 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4251 if avatar_thumbnails:
4252 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4253 if uncropped_avatar:
4254 avatar_thumbnails.append({
4255 'url': uncropped_avatar,
4256 'id': 'avatar_uncropped',
4257 'preference': 1
4258 })
4259
4260 channel_banners = self._extract_thumbnails(
4261 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4262 for banner in channel_banners:
4263 banner['preference'] = -10
4264
4265 if channel_banners:
4266 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4267 if uncropped_banner:
4268 channel_banners.append({
4269 'url': uncropped_banner,
4270 'id': 'banner_uncropped',
4271 'preference': -5
4272 })
4273
4274 primary_thumbnails = self._extract_thumbnails(
4275 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4276
4277 if playlist_id is None:
4278 playlist_id = item_id
4279
4280 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4281 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4282 if title is None:
4283 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4284 title += format_field(selected_tab, 'title', ' - %s')
4285 title += format_field(selected_tab, 'expandedText', ' - %s')
4286
4287 metadata = {
4288 'playlist_id': playlist_id,
4289 'playlist_title': title,
4290 'playlist_description': description,
4291 'uploader': channel_name,
4292 'uploader_id': channel_id,
4293 'uploader_url': channel_url,
4294 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4295 'tags': tags,
4296 'view_count': self._get_count(playlist_stats, 1),
4297 'availability': self._extract_availability(data),
4298 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4299 'playlist_count': self._get_count(playlist_stats, 0),
4300 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4301 }
4302 if not channel_id:
4303 metadata.update(self._extract_uploader(data))
4304 metadata.update({
4305 'channel': metadata['uploader'],
4306 'channel_id': metadata['uploader_id'],
4307 'channel_url': metadata['uploader_url']})
4308 return self.playlist_result(
4309 self._entries(
4310 selected_tab, playlist_id, ytcfg,
4311 self._extract_account_syncid(ytcfg, data),
4312 self._extract_visitor_data(data, ytcfg)),
4313 **metadata)
4314
4315 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4316 first_id = last_id = response = None
4317 for page_num in itertools.count(1):
4318 videos = list(self._playlist_entries(playlist))
4319 if not videos:
4320 return
4321 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4322 if start >= len(videos):
4323 return
4324 yield from videos[start:]
4325 first_id = first_id or videos[0]['id']
4326 last_id = videos[-1]['id']
4327 watch_endpoint = try_get(
4328 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4329 headers = self.generate_api_headers(
4330 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4331 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4332 query = {
4333 'playlistId': playlist_id,
4334 'videoId': watch_endpoint.get('videoId') or last_id,
4335 'index': watch_endpoint.get('index') or len(videos),
4336 'params': watch_endpoint.get('params') or 'OAE%3D'
4337 }
4338 response = self._extract_response(
4339 item_id='%s page %d' % (playlist_id, page_num),
4340 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4341 check_get_keys='contents'
4342 )
4343 playlist = try_get(
4344 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4345
4346 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4347 title = playlist.get('title') or try_get(
4348 data, lambda x: x['titleText']['simpleText'], compat_str)
4349 playlist_id = playlist.get('playlistId') or item_id
4350
4351 # Delegating everything except mix playlists to regular tab-based playlist URL
4352 playlist_url = urljoin(url, try_get(
4353 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4354 compat_str))
4355
4356 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4357 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4358 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4359
4360 if playlist_url and playlist_url != url and not is_known_unviewable:
4361 return self.url_result(
4362 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4363 video_title=title)
4364
4365 return self.playlist_result(
4366 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4367 playlist_id=playlist_id, playlist_title=title)
4368
4369 def _extract_availability(self, data):
4370 """
4371 Gets the availability of a given playlist/tab.
4372 Note: Unless YouTube tells us explicitly, we do not assume it is public
4373 @param data: response
4374 """
4375 is_private = is_unlisted = None
4376 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4377 badge_labels = self._extract_badges(renderer)
4378
4379 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4380 privacy_dropdown_entries = try_get(
4381 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4382 for renderer_dict in privacy_dropdown_entries:
4383 is_selected = try_get(
4384 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4385 if not is_selected:
4386 continue
4387 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4388 if label:
4389 badge_labels.add(label.lower())
4390 break
4391
4392 for badge_label in badge_labels:
4393 if badge_label == 'unlisted':
4394 is_unlisted = True
4395 elif badge_label == 'private':
4396 is_private = True
4397 elif badge_label == 'public':
4398 is_unlisted = is_private = False
4399 return self._availability(is_private, False, False, False, is_unlisted)
4400
4401 @staticmethod
4402 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4403 sidebar_renderer = try_get(
4404 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4405 for item in sidebar_renderer:
4406 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4407 if renderer:
4408 return renderer
4409
4410 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4411 """
4412 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4413 """
4414 browse_id = params = None
4415 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4416 if not renderer:
4417 return
4418 menu_renderer = try_get(
4419 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4420 for menu_item in menu_renderer:
4421 if not isinstance(menu_item, dict):
4422 continue
4423 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4424 text = try_get(
4425 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4426 if not text or text.lower() != 'show unavailable videos':
4427 continue
4428 browse_endpoint = try_get(
4429 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4430 browse_id = browse_endpoint.get('browseId')
4431 params = browse_endpoint.get('params')
4432 break
4433
4434 headers = self.generate_api_headers(
4435 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4436 visitor_data=self._extract_visitor_data(data, ytcfg))
4437 query = {
4438 'params': params or 'wgYCCAA=',
4439 'browseId': browse_id or 'VL%s' % item_id
4440 }
4441 return self._extract_response(
4442 item_id=item_id, headers=headers, query=query,
4443 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4444 note='Downloading API JSON with unavailable videos')
4445
4446 @functools.cached_property
4447 def skip_webpage(self):
4448 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4449
4450 def _extract_webpage(self, url, item_id, fatal=True):
4451 retries = self.get_param('extractor_retries', 3)
4452 count = -1
4453 webpage = data = last_error = None
4454 while count < retries:
4455 count += 1
4456 # Sometimes youtube returns a webpage with incomplete ytInitialData
4457 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4458 if last_error:
4459 self.report_warning('%s. Retrying ...' % last_error)
4460 try:
4461 webpage = self._download_webpage(
4462 url, item_id,
4463 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4464 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4465 except ExtractorError as e:
4466 if isinstance(e.cause, network_exceptions):
4467 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4468 last_error = error_to_compat_str(e.cause or e.msg)
4469 if count < retries:
4470 continue
4471 if fatal:
4472 raise
4473 self.report_warning(error_to_compat_str(e))
4474 break
4475 else:
4476 try:
4477 self._extract_and_report_alerts(data)
4478 except ExtractorError as e:
4479 if fatal:
4480 raise
4481 self.report_warning(error_to_compat_str(e))
4482 break
4483
4484 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4485 break
4486
4487 last_error = 'Incomplete yt initial data received'
4488 if count >= retries:
4489 if fatal:
4490 raise ExtractorError(last_error)
4491 self.report_warning(last_error)
4492 break
4493
4494 return webpage, data
4495
4496 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4497 """Use if failed to extract ytcfg (and data) from initial webpage"""
4498 if not ytcfg and self.is_authenticated:
4499 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4500 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4501 raise ExtractorError(
4502 f'{msg}. If you are not downloading private content, or '
4503 'your cookies are only for the first account and channel,'
4504 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4505 expected=True)
4506 self.report_warning(msg, only_once=True)
4507
4508 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4509 data = None
4510 if not self.skip_webpage:
4511 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4512 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4513 # Reject webpage data if redirected to home page without explicitly requesting
4514 selected_tab = self._extract_selected_tab(traverse_obj(
4515 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4516 if (url != 'https://www.youtube.com/feed/recommended'
4517 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4518 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4519 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4520 if fatal:
4521 raise ExtractorError(msg, expected=True)
4522 self.report_warning(msg, only_once=True)
4523 if not data:
4524 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4525 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4526 return data, ytcfg
4527
4528 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4529 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4530 resolve_response = self._extract_response(
4531 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4532 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4533 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4534 for ep_key, ep in endpoints.items():
4535 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4536 if params:
4537 return self._extract_response(
4538 item_id=item_id, query=params, ep=ep, headers=headers,
4539 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4540 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4541 err_note = 'Failed to resolve url (does the playlist exist?)'
4542 if fatal:
4543 raise ExtractorError(err_note, expected=True)
4544 self.report_warning(err_note, item_id)
4545
4546 _SEARCH_PARAMS = None
4547
4548 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4549 data = {'query': query}
4550 if params is NO_DEFAULT:
4551 params = self._SEARCH_PARAMS
4552 if params:
4553 data['params'] = params
4554
4555 content_keys = (
4556 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4557 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4558 # ytmusic search
4559 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4560 ('continuationContents', ),
4561 )
4562 display_id = f'query "{query}"'
4563 check_get_keys = tuple({keys[0] for keys in content_keys})
4564 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4565 self._report_playlist_authcheck(ytcfg, fatal=False)
4566
4567 continuation_list = [None]
4568 search = None
4569 for page_num in itertools.count(1):
4570 data.update(continuation_list[0] or {})
4571 headers = self.generate_api_headers(
4572 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4573 search = self._extract_response(
4574 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4575 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4576 slr_contents = traverse_obj(search, *content_keys)
4577 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4578 if not continuation_list[0]:
4579 break
4580
4581
4582class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4583 IE_DESC = 'YouTube Tabs'
4584 _VALID_URL = r'''(?x:
4585 https?://
4586 (?:\w+\.)?
4587 (?:
4588 youtube(?:kids)?\.com|
4589 %(invidious)s
4590 )/
4591 (?:
4592 (?P<channel_type>channel|c|user|browse)/|
4593 (?P<not_channel>
4594 feed/|hashtag/|
4595 (?:playlist|watch)\?.*?\blist=
4596 )|
4597 (?!(?:%(reserved_names)s)\b) # Direct URLs
4598 )
4599 (?P<id>[^/?\#&]+)
4600 )''' % {
4601 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4602 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4603 }
4604 IE_NAME = 'youtube:tab'
4605
4606 _TESTS = [{
4607 'note': 'playlists, multipage',
4608 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4609 'playlist_mincount': 94,
4610 'info_dict': {
4611 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4612 'title': 'Igor Kleiner - Playlists',
4613 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4614 'uploader': 'Igor Kleiner',
4615 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4616 'channel': 'Igor Kleiner',
4617 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4618 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4619 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4620 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4621 'channel_follower_count': int
4622 },
4623 }, {
4624 'note': 'playlists, multipage, different order',
4625 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4626 'playlist_mincount': 94,
4627 'info_dict': {
4628 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4629 'title': 'Igor Kleiner - Playlists',
4630 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4631 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4632 'uploader': 'Igor Kleiner',
4633 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4634 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4635 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4636 'channel': 'Igor Kleiner',
4637 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4638 'channel_follower_count': int
4639 },
4640 }, {
4641 'note': 'playlists, series',
4642 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4643 'playlist_mincount': 5,
4644 'info_dict': {
4645 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4646 'title': '3Blue1Brown - Playlists',
4647 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4648 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4649 'uploader': '3Blue1Brown',
4650 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4651 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4652 'channel': '3Blue1Brown',
4653 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4654 'tags': ['Mathematics'],
4655 'channel_follower_count': int
4656 },
4657 }, {
4658 'note': 'playlists, singlepage',
4659 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4660 'playlist_mincount': 4,
4661 'info_dict': {
4662 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4663 'title': 'ThirstForScience - Playlists',
4664 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4665 'uploader': 'ThirstForScience',
4666 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4667 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4668 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4669 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4670 'tags': 'count:13',
4671 'channel': 'ThirstForScience',
4672 'channel_follower_count': int
4673 }
4674 }, {
4675 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4676 'only_matching': True,
4677 }, {
4678 'note': 'basic, single video playlist',
4679 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4680 'info_dict': {
4681 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4682 'uploader': 'Sergey M.',
4683 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4684 'title': 'youtube-dl public playlist',
4685 'description': '',
4686 'tags': [],
4687 'view_count': int,
4688 'modified_date': '20201130',
4689 'channel': 'Sergey M.',
4690 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4691 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4692 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4693 },
4694 'playlist_count': 1,
4695 }, {
4696 'note': 'empty playlist',
4697 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4698 'info_dict': {
4699 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4700 'uploader': 'Sergey M.',
4701 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4702 'title': 'youtube-dl empty playlist',
4703 'tags': [],
4704 'channel': 'Sergey M.',
4705 'description': '',
4706 'modified_date': '20160902',
4707 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4708 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4709 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4710 },
4711 'playlist_count': 0,
4712 }, {
4713 'note': 'Home tab',
4714 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4715 'info_dict': {
4716 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4717 'title': 'lex will - Home',
4718 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4719 'uploader': 'lex will',
4720 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4721 'channel': 'lex will',
4722 'tags': ['bible', 'history', 'prophesy'],
4723 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4724 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4725 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4726 'channel_follower_count': int
4727 },
4728 'playlist_mincount': 2,
4729 }, {
4730 'note': 'Videos tab',
4731 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4732 'info_dict': {
4733 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4734 'title': 'lex will - Videos',
4735 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4736 'uploader': 'lex will',
4737 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4738 'tags': ['bible', 'history', 'prophesy'],
4739 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4740 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4742 'channel': 'lex will',
4743 'channel_follower_count': int
4744 },
4745 'playlist_mincount': 975,
4746 }, {
4747 'note': 'Videos tab, sorted by popular',
4748 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4749 'info_dict': {
4750 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4751 'title': 'lex will - Videos',
4752 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4753 'uploader': 'lex will',
4754 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4755 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4756 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4757 'channel': 'lex will',
4758 'tags': ['bible', 'history', 'prophesy'],
4759 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4760 'channel_follower_count': int
4761 },
4762 'playlist_mincount': 199,
4763 }, {
4764 'note': 'Playlists tab',
4765 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4766 'info_dict': {
4767 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4768 'title': 'lex will - Playlists',
4769 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4770 'uploader': 'lex will',
4771 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4772 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4773 'channel': 'lex will',
4774 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4775 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4776 'tags': ['bible', 'history', 'prophesy'],
4777 'channel_follower_count': int
4778 },
4779 'playlist_mincount': 17,
4780 }, {
4781 'note': 'Community tab',
4782 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4783 'info_dict': {
4784 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4785 'title': 'lex will - Community',
4786 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4787 'uploader': 'lex will',
4788 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4789 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4790 'channel': 'lex will',
4791 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4792 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4793 'tags': ['bible', 'history', 'prophesy'],
4794 'channel_follower_count': int
4795 },
4796 'playlist_mincount': 18,
4797 }, {
4798 'note': 'Channels tab',
4799 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4800 'info_dict': {
4801 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4802 'title': 'lex will - Channels',
4803 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4804 'uploader': 'lex will',
4805 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4806 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4807 'channel': 'lex will',
4808 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4809 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4810 'tags': ['bible', 'history', 'prophesy'],
4811 'channel_follower_count': int
4812 },
4813 'playlist_mincount': 12,
4814 }, {
4815 'note': 'Search tab',
4816 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4817 'playlist_mincount': 40,
4818 'info_dict': {
4819 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4820 'title': '3Blue1Brown - Search - linear algebra',
4821 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4822 'uploader': '3Blue1Brown',
4823 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4824 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4825 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4826 'tags': ['Mathematics'],
4827 'channel': '3Blue1Brown',
4828 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4829 'channel_follower_count': int
4830 },
4831 }, {
4832 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4833 'only_matching': True,
4834 }, {
4835 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4836 'only_matching': True,
4837 }, {
4838 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4839 'only_matching': True,
4840 }, {
4841 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4842 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4843 'info_dict': {
4844 'title': '29C3: Not my department',
4845 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4846 'uploader': 'Christiaan008',
4847 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4848 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4849 'tags': [],
4850 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4851 'view_count': int,
4852 'modified_date': '20150605',
4853 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4854 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4855 'channel': 'Christiaan008',
4856 },
4857 'playlist_count': 96,
4858 }, {
4859 'note': 'Large playlist',
4860 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4861 'info_dict': {
4862 'title': 'Uploads from Cauchemar',
4863 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4864 'uploader': 'Cauchemar',
4865 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4866 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4867 'tags': [],
4868 'modified_date': r're:\d{8}',
4869 'channel': 'Cauchemar',
4870 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4871 'view_count': int,
4872 'description': '',
4873 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4874 },
4875 'playlist_mincount': 1123,
4876 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4877 }, {
4878 'note': 'even larger playlist, 8832 videos',
4879 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4880 'only_matching': True,
4881 }, {
4882 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4883 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4884 'info_dict': {
4885 'title': 'Uploads from Interstellar Movie',
4886 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4887 'uploader': 'Interstellar Movie',
4888 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4889 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4890 'tags': [],
4891 'view_count': int,
4892 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4893 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4894 'channel': 'Interstellar Movie',
4895 'description': '',
4896 'modified_date': r're:\d{8}',
4897 },
4898 'playlist_mincount': 21,
4899 }, {
4900 'note': 'Playlist with "show unavailable videos" button',
4901 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4902 'info_dict': {
4903 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4904 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4905 'uploader': 'Phim Siêu Nhân Nhật Bản',
4906 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4907 'view_count': int,
4908 'channel': 'Phim Siêu Nhân Nhật Bản',
4909 'tags': [],
4910 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4911 'description': '',
4912 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4913 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4914 'modified_date': r're:\d{8}',
4915 },
4916 'playlist_mincount': 200,
4917 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4918 }, {
4919 'note': 'Playlist with unavailable videos in page 7',
4920 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4921 'info_dict': {
4922 'title': 'Uploads from BlankTV',
4923 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4924 'uploader': 'BlankTV',
4925 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4926 'channel': 'BlankTV',
4927 'channel_url': 'https://www.youtube.com/c/blanktv',
4928 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4929 'view_count': int,
4930 'tags': [],
4931 'uploader_url': 'https://www.youtube.com/c/blanktv',
4932 'modified_date': r're:\d{8}',
4933 'description': '',
4934 },
4935 'playlist_mincount': 1000,
4936 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4937 }, {
4938 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4939 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4940 'info_dict': {
4941 'title': 'Data Analysis with Dr Mike Pound',
4942 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4943 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4944 'uploader': 'Computerphile',
4945 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4946 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4947 'tags': [],
4948 'view_count': int,
4949 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4950 'channel_url': 'https://www.youtube.com/user/Computerphile',
4951 'channel': 'Computerphile',
4952 },
4953 'playlist_mincount': 11,
4954 }, {
4955 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4956 'only_matching': True,
4957 }, {
4958 'note': 'Playlist URL that does not actually serve a playlist',
4959 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4960 'info_dict': {
4961 'id': 'FqZTN594JQw',
4962 'ext': 'webm',
4963 'title': "Smiley's People 01 detective, Adventure Series, Action",
4964 'uploader': 'STREEM',
4965 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4966 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4967 'upload_date': '20150526',
4968 'license': 'Standard YouTube License',
4969 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4970 'categories': ['People & Blogs'],
4971 'tags': list,
4972 'view_count': int,
4973 'like_count': int,
4974 },
4975 'params': {
4976 'skip_download': True,
4977 },
4978 'skip': 'This video is not available.',
4979 'add_ie': [YoutubeIE.ie_key()],
4980 }, {
4981 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4982 'only_matching': True,
4983 }, {
4984 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4985 'only_matching': True,
4986 }, {
4987 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4988 'info_dict': {
4989 'id': 'GgL890LIznQ', # This will keep changing
4990 'ext': 'mp4',
4991 'title': str,
4992 'uploader': 'Sky News',
4993 'uploader_id': 'skynews',
4994 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4995 'upload_date': r're:\d{8}',
4996 'description': str,
4997 'categories': ['News & Politics'],
4998 'tags': list,
4999 'like_count': int,
5000 'release_timestamp': 1642502819,
5001 'channel': 'Sky News',
5002 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5003 'age_limit': 0,
5004 'view_count': int,
5005 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
5006 'playable_in_embed': True,
5007 'release_date': '20220118',
5008 'availability': 'public',
5009 'live_status': 'is_live',
5010 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5011 'channel_follower_count': int
5012 },
5013 'params': {
5014 'skip_download': True,
5015 },
5016 'expected_warnings': ['Ignoring subtitle tracks found in '],
5017 }, {
5018 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5019 'info_dict': {
5020 'id': 'a48o2S1cPoo',
5021 'ext': 'mp4',
5022 'title': 'The Young Turks - Live Main Show',
5023 'uploader': 'The Young Turks',
5024 'uploader_id': 'TheYoungTurks',
5025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
5026 'upload_date': '20150715',
5027 'license': 'Standard YouTube License',
5028 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5029 'categories': ['News & Politics'],
5030 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5031 'like_count': int,
5032 },
5033 'params': {
5034 'skip_download': True,
5035 },
5036 'only_matching': True,
5037 }, {
5038 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5039 'only_matching': True,
5040 }, {
5041 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5042 'only_matching': True,
5043 }, {
5044 'note': 'A channel that is not live. Should raise error',
5045 'url': 'https://www.youtube.com/user/numberphile/live',
5046 'only_matching': True,
5047 }, {
5048 'url': 'https://www.youtube.com/feed/trending',
5049 'only_matching': True,
5050 }, {
5051 'url': 'https://www.youtube.com/feed/library',
5052 'only_matching': True,
5053 }, {
5054 'url': 'https://www.youtube.com/feed/history',
5055 'only_matching': True,
5056 }, {
5057 'url': 'https://www.youtube.com/feed/subscriptions',
5058 'only_matching': True,
5059 }, {
5060 'url': 'https://www.youtube.com/feed/watch_later',
5061 'only_matching': True,
5062 }, {
5063 'note': 'Recommended - redirects to home page.',
5064 'url': 'https://www.youtube.com/feed/recommended',
5065 'only_matching': True,
5066 }, {
5067 'note': 'inline playlist with not always working continuations',
5068 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5069 'only_matching': True,
5070 }, {
5071 'url': 'https://www.youtube.com/course',
5072 'only_matching': True,
5073 }, {
5074 'url': 'https://www.youtube.com/zsecurity',
5075 'only_matching': True,
5076 }, {
5077 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5078 'only_matching': True,
5079 }, {
5080 'url': 'https://www.youtube.com/TheYoungTurks/live',
5081 'only_matching': True,
5082 }, {
5083 'url': 'https://www.youtube.com/hashtag/cctv9',
5084 'info_dict': {
5085 'id': 'cctv9',
5086 'title': '#cctv9',
5087 'tags': [],
5088 },
5089 'playlist_mincount': 350,
5090 }, {
5091 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5092 'only_matching': True,
5093 }, {
5094 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5095 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5096 'only_matching': True
5097 }, {
5098 'note': '/browse/ should redirect to /channel/',
5099 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5100 'only_matching': True
5101 }, {
5102 'note': 'VLPL, should redirect to playlist?list=PL...',
5103 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5104 'info_dict': {
5105 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5106 'uploader': 'NoCopyrightSounds',
5107 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5108 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5109 'title': 'NCS Releases',
5110 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5111 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5112 'modified_date': r're:\d{8}',
5113 'view_count': int,
5114 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5115 'tags': [],
5116 'channel': 'NoCopyrightSounds',
5117 },
5118 'playlist_mincount': 166,
5119 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5120 }, {
5121 'note': 'Topic, should redirect to playlist?list=UU...',
5122 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5123 'info_dict': {
5124 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5125 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5126 'title': 'Uploads from Royalty Free Music - Topic',
5127 'uploader': 'Royalty Free Music - Topic',
5128 'tags': [],
5129 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5130 'channel': 'Royalty Free Music - Topic',
5131 'view_count': int,
5132 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5133 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5134 'modified_date': r're:\d{8}',
5135 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5136 'description': '',
5137 },
5138 'expected_warnings': [
5139 'The URL does not have a videos tab',
5140 r'[Uu]navailable videos (are|will be) hidden',
5141 ],
5142 'playlist_mincount': 101,
5143 }, {
5144 'note': 'Topic without a UU playlist',
5145 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5146 'info_dict': {
5147 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5148 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5149 'tags': [],
5150 },
5151 'expected_warnings': [
5152 'the playlist redirect gave error',
5153 ],
5154 'playlist_mincount': 9,
5155 }, {
5156 'note': 'Youtube music Album',
5157 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5158 'info_dict': {
5159 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5160 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5161 'tags': [],
5162 'view_count': int,
5163 'description': '',
5164 'availability': 'unlisted',
5165 'modified_date': r're:\d{8}',
5166 },
5167 'playlist_count': 50,
5168 }, {
5169 'note': 'unlisted single video playlist',
5170 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5171 'info_dict': {
5172 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5173 'uploader': 'colethedj',
5174 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5175 'title': 'yt-dlp unlisted playlist test',
5176 'availability': 'unlisted',
5177 'tags': [],
5178 'modified_date': '20211208',
5179 'channel': 'colethedj',
5180 'view_count': int,
5181 'description': '',
5182 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5183 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5184 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5185 },
5186 'playlist_count': 1,
5187 }, {
5188 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5189 'url': 'https://www.youtube.com/feed/recommended',
5190 'info_dict': {
5191 'id': 'recommended',
5192 'title': 'recommended',
5193 'tags': [],
5194 },
5195 'playlist_mincount': 50,
5196 'params': {
5197 'skip_download': True,
5198 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5199 },
5200 }, {
5201 'note': 'API Fallback: /videos tab, sorted by oldest first',
5202 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5203 'info_dict': {
5204 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5205 'title': 'Cody\'sLab - Videos',
5206 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5207 'uploader': 'Cody\'sLab',
5208 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5209 'channel': 'Cody\'sLab',
5210 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5211 'tags': [],
5212 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5213 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5214 'channel_follower_count': int
5215 },
5216 'playlist_mincount': 650,
5217 'params': {
5218 'skip_download': True,
5219 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5220 },
5221 }, {
5222 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5223 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5224 'info_dict': {
5225 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5226 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5227 'title': 'Uploads from Royalty Free Music - Topic',
5228 'uploader': 'Royalty Free Music - Topic',
5229 'modified_date': r're:\d{8}',
5230 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5231 'description': '',
5232 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5233 'tags': [],
5234 'channel': 'Royalty Free Music - Topic',
5235 'view_count': int,
5236 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5237 },
5238 'expected_warnings': [
5239 'does not have a videos tab',
5240 r'[Uu]navailable videos (are|will be) hidden',
5241 ],
5242 'playlist_mincount': 101,
5243 'params': {
5244 'skip_download': True,
5245 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5246 },
5247 }, {
5248 'note': 'non-standard redirect to regional channel',
5249 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5250 'only_matching': True
5251 }, {
5252 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5253 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5254 'info_dict': {
5255 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5256 'modified_date': '20220407',
5257 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5258 'tags': [],
5259 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5260 'uploader': 'pukkandan',
5261 'availability': 'unlisted',
5262 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5263 'channel': 'pukkandan',
5264 'description': 'Test for collaborative playlist',
5265 'title': 'yt-dlp test - collaborative playlist',
5266 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5267 },
5268 'playlist_mincount': 2
5269 }]
5270
5271 @classmethod
5272 def suitable(cls, url):
5273 return False if YoutubeIE.suitable(url) else super().suitable(url)
5274
5275 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5276
5277 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5278 def _real_extract(self, url, smuggled_data):
5279 item_id = self._match_id(url)
5280 url = compat_urlparse.urlunparse(
5281 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5282 compat_opts = self.get_param('compat_opts', [])
5283
5284 def get_mobj(url):
5285 mobj = self._URL_RE.match(url).groupdict()
5286 mobj.update((k, '') for k, v in mobj.items() if v is None)
5287 return mobj
5288
5289 mobj, redirect_warning = get_mobj(url), None
5290 # Youtube returns incomplete data if tabname is not lower case
5291 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5292 if is_channel:
5293 if smuggled_data.get('is_music_url'):
5294 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5295 item_id = item_id[2:]
5296 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5297 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5298 mdata = self._extract_tab_endpoint(
5299 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5300 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5301 get_all=False, expected_type=compat_str)
5302 if not murl:
5303 raise ExtractorError('Failed to resolve album to playlist')
5304 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5305 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5306 pre = f'https://www.youtube.com/channel/{item_id}'
5307
5308 original_tab_name = tab
5309 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5310 # Home URLs should redirect to /videos/
5311 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5312 'To download only the videos in the home page, add a "/featured" to the URL')
5313 tab = '/videos'
5314
5315 url = ''.join((pre, tab, post))
5316 mobj = get_mobj(url)
5317
5318 # Handle both video/playlist URLs
5319 qs = parse_qs(url)
5320 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5321
5322 if not video_id and mobj['not_channel'].startswith('watch'):
5323 if not playlist_id:
5324 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5325 raise ExtractorError('Unable to recognize tab page')
5326 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5327 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5328 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5329 mobj = get_mobj(url)
5330
5331 if video_id and playlist_id:
5332 if self.get_param('noplaylist'):
5333 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5334 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5335 ie=YoutubeIE.ie_key(), video_id=video_id)
5336 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5337
5338 data, ytcfg = self._extract_data(url, item_id)
5339
5340 # YouTube may provide a non-standard redirect to the regional channel
5341 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5342 redirect_url = traverse_obj(
5343 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5344 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5345 redirect_url = ''.join((
5346 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5347 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5348 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5349
5350 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5351 if tabs:
5352 selected_tab = self._extract_selected_tab(tabs)
5353 selected_tab_name = selected_tab.get('title', '').lower()
5354 if selected_tab_name == 'home':
5355 selected_tab_name = 'featured'
5356 requested_tab_name = mobj['tab'][1:]
5357 if 'no-youtube-channel-redirect' not in compat_opts:
5358 if requested_tab_name == 'live':
5359 # Live tab should have redirected to the video
5360 raise ExtractorError('The channel is not currently live', expected=True)
5361 if requested_tab_name not in ('', selected_tab_name):
5362 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5363 if not original_tab_name:
5364 if item_id[:2] == 'UC':
5365 # Topic channels don't have /videos. Use the equivalent playlist instead
5366 pl_id = f'UU{item_id[2:]}'
5367 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5368 try:
5369 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5370 except ExtractorError:
5371 redirect_warning += ' and the playlist redirect gave error'
5372 else:
5373 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5374 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5375 if selected_tab_name and selected_tab_name != requested_tab_name:
5376 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5377 else:
5378 raise ExtractorError(redirect_warning, expected=True)
5379
5380 if redirect_warning:
5381 self.to_screen(redirect_warning)
5382 self.write_debug(f'Final URL: {url}')
5383
5384 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5385 if 'no-youtube-unavailable-videos' not in compat_opts:
5386 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5387 self._extract_and_report_alerts(data, only_once=True)
5388 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5389 if tabs:
5390 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5391
5392 playlist = traverse_obj(
5393 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5394 if playlist:
5395 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5396
5397 video_id = traverse_obj(
5398 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5399 if video_id:
5400 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5401 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5402 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5403 ie=YoutubeIE.ie_key(), video_id=video_id)
5404
5405 raise ExtractorError('Unable to recognize tab page')
5406
5407
5408class YoutubePlaylistIE(InfoExtractor):
5409 IE_DESC = 'YouTube playlists'
5410 _VALID_URL = r'''(?x)(?:
5411 (?:https?://)?
5412 (?:\w+\.)?
5413 (?:
5414 (?:
5415 youtube(?:kids)?\.com|
5416 %(invidious)s
5417 )
5418 /.*?\?.*?\blist=
5419 )?
5420 (?P<id>%(playlist_id)s)
5421 )''' % {
5422 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5423 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5424 }
5425 IE_NAME = 'youtube:playlist'
5426 _TESTS = [{
5427 'note': 'issue #673',
5428 'url': 'PLBB231211A4F62143',
5429 'info_dict': {
5430 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5431 'id': 'PLBB231211A4F62143',
5432 'uploader': 'Wickman',
5433 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5434 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5435 'view_count': int,
5436 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5437 'modified_date': r're:\d{8}',
5438 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5439 'channel': 'Wickman',
5440 'tags': [],
5441 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5442 },
5443 'playlist_mincount': 29,
5444 }, {
5445 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5446 'info_dict': {
5447 'title': 'YDL_safe_search',
5448 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5449 },
5450 'playlist_count': 2,
5451 'skip': 'This playlist is private',
5452 }, {
5453 'note': 'embedded',
5454 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5455 'playlist_count': 4,
5456 'info_dict': {
5457 'title': 'JODA15',
5458 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5459 'uploader': 'milan',
5460 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5461 'description': '',
5462 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5463 'tags': [],
5464 'modified_date': '20140919',
5465 'view_count': int,
5466 'channel': 'milan',
5467 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5468 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5469 },
5470 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5471 }, {
5472 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5473 'playlist_mincount': 654,
5474 'info_dict': {
5475 'title': '2018 Chinese New Singles (11/6 updated)',
5476 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5477 'uploader': 'LBK',
5478 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5479 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5480 'channel': 'LBK',
5481 'view_count': int,
5482 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5483 'tags': [],
5484 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5485 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5486 'modified_date': r're:\d{8}',
5487 },
5488 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5489 }, {
5490 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5491 'only_matching': True,
5492 }, {
5493 # music album playlist
5494 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5495 'only_matching': True,
5496 }]
5497
5498 @classmethod
5499 def suitable(cls, url):
5500 if YoutubeTabIE.suitable(url):
5501 return False
5502 from ..utils import parse_qs
5503 qs = parse_qs(url)
5504 if qs.get('v', [None])[0]:
5505 return False
5506 return super().suitable(url)
5507
5508 def _real_extract(self, url):
5509 playlist_id = self._match_id(url)
5510 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5511 url = update_url_query(
5512 'https://www.youtube.com/playlist',
5513 parse_qs(url) or {'list': playlist_id})
5514 if is_music_url:
5515 url = smuggle_url(url, {'is_music_url': True})
5516 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5517
5518
5519class YoutubeYtBeIE(InfoExtractor):
5520 IE_DESC = 'youtu.be'
5521 _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}
5522 _TESTS = [{
5523 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5524 'info_dict': {
5525 'id': 'yeWKywCrFtk',
5526 'ext': 'mp4',
5527 'title': 'Small Scale Baler and Braiding Rugs',
5528 'uploader': 'Backus-Page House Museum',
5529 'uploader_id': 'backuspagemuseum',
5530 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5531 'upload_date': '20161008',
5532 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5533 'categories': ['Nonprofits & Activism'],
5534 'tags': list,
5535 'like_count': int,
5536 'age_limit': 0,
5537 'playable_in_embed': True,
5538 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5539 'channel': 'Backus-Page House Museum',
5540 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5541 'live_status': 'not_live',
5542 'view_count': int,
5543 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5544 'availability': 'public',
5545 'duration': 59,
5546 },
5547 'params': {
5548 'noplaylist': True,
5549 'skip_download': True,
5550 },
5551 }, {
5552 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5553 'only_matching': True,
5554 }]
5555
5556 def _real_extract(self, url):
5557 mobj = self._match_valid_url(url)
5558 video_id = mobj.group('id')
5559 playlist_id = mobj.group('playlist_id')
5560 return self.url_result(
5561 update_url_query('https://www.youtube.com/watch', {
5562 'v': video_id,
5563 'list': playlist_id,
5564 'feature': 'youtu.be',
5565 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5566
5567
5568class YoutubeLivestreamEmbedIE(InfoExtractor):
5569 IE_DESC = 'YouTube livestream embeds'
5570 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5571 _TESTS = [{
5572 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5573 'only_matching': True,
5574 }]
5575
5576 def _real_extract(self, url):
5577 channel_id = self._match_id(url)
5578 return self.url_result(
5579 f'https://www.youtube.com/channel/{channel_id}/live',
5580 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5581
5582
5583class YoutubeYtUserIE(InfoExtractor):
5584 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5585 IE_NAME = 'youtube:user'
5586 _VALID_URL = r'ytuser:(?P<id>.+)'
5587 _TESTS = [{
5588 'url': 'ytuser:phihag',
5589 'only_matching': True,
5590 }]
5591
5592 def _real_extract(self, url):
5593 user_id = self._match_id(url)
5594 return self.url_result(
5595 'https://www.youtube.com/user/%s/videos' % user_id,
5596 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5597
5598
5599class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5600 IE_NAME = 'youtube:favorites'
5601 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5602 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5603 _LOGIN_REQUIRED = True
5604 _TESTS = [{
5605 'url': ':ytfav',
5606 'only_matching': True,
5607 }, {
5608 'url': ':ytfavorites',
5609 'only_matching': True,
5610 }]
5611
5612 def _real_extract(self, url):
5613 return self.url_result(
5614 'https://www.youtube.com/playlist?list=LL',
5615 ie=YoutubeTabIE.ie_key())
5616
5617
5618class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5619 IE_NAME = 'youtube:notif'
5620 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5621 _VALID_URL = r':ytnotif(?:ication)?s?'
5622 _LOGIN_REQUIRED = True
5623 _TESTS = [{
5624 'url': ':ytnotif',
5625 'only_matching': True,
5626 }, {
5627 'url': ':ytnotifications',
5628 'only_matching': True,
5629 }]
5630
5631 def _extract_notification_menu(self, response, continuation_list):
5632 notification_list = traverse_obj(
5633 response,
5634 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5635 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5636 expected_type=list) or []
5637 continuation_list[0] = None
5638 for item in notification_list:
5639 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5640 if entry:
5641 yield entry
5642 continuation = item.get('continuationItemRenderer')
5643 if continuation:
5644 continuation_list[0] = continuation
5645
5646 def _extract_notification_renderer(self, notification):
5647 video_id = traverse_obj(
5648 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5649 url = f'https://www.youtube.com/watch?v={video_id}'
5650 channel_id = None
5651 if not video_id:
5652 browse_ep = traverse_obj(
5653 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5654 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5655 post_id = self._search_regex(
5656 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5657 'post id', default=None)
5658 if not channel_id or not post_id:
5659 return
5660 # The direct /post url redirects to this in the browser
5661 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5662
5663 channel = traverse_obj(
5664 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5665 expected_type=str)
5666 notification_title = self._get_text(notification, 'shortMessage')
5667 if notification_title:
5668 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
5669 # TODO: handle recommended videos
5670 title = self._search_regex(
5671 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
5672 'video title', default=None)
5673 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5674 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5675 else None)
5676 return {
5677 '_type': 'url',
5678 'url': url,
5679 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5680 'video_id': video_id,
5681 'title': title,
5682 'channel_id': channel_id,
5683 'channel': channel,
5684 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5685 'upload_date': upload_date,
5686 }
5687
5688 def _notification_menu_entries(self, ytcfg):
5689 continuation_list = [None]
5690 response = None
5691 for page in itertools.count(1):
5692 ctoken = traverse_obj(
5693 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5694 response = self._extract_response(
5695 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5696 ep='notification/get_notification_menu', check_get_keys='actions',
5697 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5698 yield from self._extract_notification_menu(response, continuation_list)
5699 if not continuation_list[0]:
5700 break
5701
5702 def _real_extract(self, url):
5703 display_id = 'notifications'
5704 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5705 self._report_playlist_authcheck(ytcfg)
5706 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5707
5708
5709class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5710 IE_DESC = 'YouTube search'
5711 IE_NAME = 'youtube:search'
5712 _SEARCH_KEY = 'ytsearch'
5713 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5714 _TESTS = [{
5715 'url': 'ytsearch5:youtube-dl test video',
5716 'playlist_count': 5,
5717 'info_dict': {
5718 'id': 'youtube-dl test video',
5719 'title': 'youtube-dl test video',
5720 }
5721 }]
5722
5723
5724class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5725 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5726 _SEARCH_KEY = 'ytsearchdate'
5727 IE_DESC = 'YouTube search, newest videos first'
5728 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5729 _TESTS = [{
5730 'url': 'ytsearchdate5:youtube-dl test video',
5731 'playlist_count': 5,
5732 'info_dict': {
5733 'id': 'youtube-dl test video',
5734 'title': 'youtube-dl test video',
5735 }
5736 }]
5737
5738
5739class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5740 IE_DESC = 'YouTube search URLs with sorting and filter support'
5741 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5742 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5743 _TESTS = [{
5744 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5745 'playlist_mincount': 5,
5746 'info_dict': {
5747 'id': 'youtube-dl test video',
5748 'title': 'youtube-dl test video',
5749 }
5750 }, {
5751 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5752 'playlist_mincount': 5,
5753 'info_dict': {
5754 'id': 'python',
5755 'title': 'python',
5756 }
5757 }, {
5758 'url': 'https://www.youtube.com/results?search_query=%23cats',
5759 'playlist_mincount': 1,
5760 'info_dict': {
5761 'id': '#cats',
5762 'title': '#cats',
5763 'entries': [{
5764 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5765 'title': '#cats',
5766 }],
5767 },
5768 }, {
5769 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5770 'only_matching': True,
5771 }]
5772
5773 def _real_extract(self, url):
5774 qs = parse_qs(url)
5775 query = (qs.get('search_query') or qs.get('q'))[0]
5776 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5777
5778
5779class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5780 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5781 IE_NAME = 'youtube:music:search_url'
5782 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5783 _TESTS = [{
5784 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5785 'playlist_count': 16,
5786 'info_dict': {
5787 'id': 'royalty free music',
5788 'title': 'royalty free music',
5789 }
5790 }, {
5791 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5792 'playlist_mincount': 30,
5793 'info_dict': {
5794 'id': 'royalty free music - songs',
5795 'title': 'royalty free music - songs',
5796 },
5797 'params': {'extract_flat': 'in_playlist'}
5798 }, {
5799 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5800 'playlist_mincount': 30,
5801 'info_dict': {
5802 'id': 'royalty free music - community playlists',
5803 'title': 'royalty free music - community playlists',
5804 },
5805 'params': {'extract_flat': 'in_playlist'}
5806 }]
5807
5808 _SECTIONS = {
5809 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5810 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5811 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5812 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5813 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5814 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5815 }
5816
5817 def _real_extract(self, url):
5818 qs = parse_qs(url)
5819 query = (qs.get('search_query') or qs.get('q'))[0]
5820 params = qs.get('sp', (None,))[0]
5821 if params:
5822 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5823 else:
5824 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5825 params = self._SECTIONS.get(section)
5826 if not params:
5827 section = None
5828 title = join_nonempty(query, section, delim=' - ')
5829 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5830
5831
5832class YoutubeFeedsInfoExtractor(InfoExtractor):
5833 """
5834 Base class for feed extractors
5835 Subclasses must re-define the _FEED_NAME property.
5836 """
5837 _LOGIN_REQUIRED = True
5838 _FEED_NAME = 'feeds'
5839
5840 def _real_initialize(self):
5841 YoutubeBaseInfoExtractor._check_login_required(self)
5842
5843 @classproperty
5844 def IE_NAME(self):
5845 return f'youtube:{self._FEED_NAME}'
5846
5847 def _real_extract(self, url):
5848 return self.url_result(
5849 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5850
5851
5852class YoutubeWatchLaterIE(InfoExtractor):
5853 IE_NAME = 'youtube:watchlater'
5854 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5855 _VALID_URL = r':ytwatchlater'
5856 _TESTS = [{
5857 'url': ':ytwatchlater',
5858 'only_matching': True,
5859 }]
5860
5861 def _real_extract(self, url):
5862 return self.url_result(
5863 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5864
5865
5866class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5867 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5868 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5869 _FEED_NAME = 'recommended'
5870 _LOGIN_REQUIRED = False
5871 _TESTS = [{
5872 'url': ':ytrec',
5873 'only_matching': True,
5874 }, {
5875 'url': ':ytrecommended',
5876 'only_matching': True,
5877 }, {
5878 'url': 'https://youtube.com',
5879 'only_matching': True,
5880 }]
5881
5882
5883class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5884 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5885 _VALID_URL = r':ytsub(?:scription)?s?'
5886 _FEED_NAME = 'subscriptions'
5887 _TESTS = [{
5888 'url': ':ytsubs',
5889 'only_matching': True,
5890 }, {
5891 'url': ':ytsubscriptions',
5892 'only_matching': True,
5893 }]
5894
5895
5896class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5897 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5898 _VALID_URL = r':ythis(?:tory)?'
5899 _FEED_NAME = 'history'
5900 _TESTS = [{
5901 'url': ':ythistory',
5902 'only_matching': True,
5903 }]
5904
5905
5906class YoutubeStoriesIE(InfoExtractor):
5907 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5908 IE_NAME = 'youtube:stories'
5909 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5910 _TESTS = [{
5911 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5912 'only_matching': True,
5913 }]
5914
5915 def _real_extract(self, url):
5916 playlist_id = f'RLTD{self._match_id(url)}'
5917 return self.url_result(
5918 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5919 ie=YoutubeTabIE, video_id=playlist_id)
5920
5921
5922class YoutubeTruncatedURLIE(InfoExtractor):
5923 IE_NAME = 'youtube:truncated_url'
5924 IE_DESC = False # Do not list
5925 _VALID_URL = r'''(?x)
5926 (?:https?://)?
5927 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5928 (?:watch\?(?:
5929 feature=[a-z_]+|
5930 annotation_id=annotation_[^&]+|
5931 x-yt-cl=[0-9]+|
5932 hl=[^&]*|
5933 t=[0-9]+
5934 )?
5935 |
5936 attribution_link\?a=[^&]+
5937 )
5938 $
5939 '''
5940
5941 _TESTS = [{
5942 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5943 'only_matching': True,
5944 }, {
5945 'url': 'https://www.youtube.com/watch?',
5946 'only_matching': True,
5947 }, {
5948 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5949 'only_matching': True,
5950 }, {
5951 'url': 'https://www.youtube.com/watch?feature=foo',
5952 'only_matching': True,
5953 }, {
5954 'url': 'https://www.youtube.com/watch?hl=en-GB',
5955 'only_matching': True,
5956 }, {
5957 'url': 'https://www.youtube.com/watch?t=2372',
5958 'only_matching': True,
5959 }]
5960
5961 def _real_extract(self, url):
5962 raise ExtractorError(
5963 'Did you forget to quote the URL? Remember that & is a meta '
5964 'character in most shells, so you want to put the URL in quotes, '
5965 'like youtube-dl '
5966 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5967 ' or simply youtube-dl BaW_jenozKc .',
5968 expected=True)
5969
5970
5971class YoutubeClipIE(InfoExtractor):
5972 IE_NAME = 'youtube:clip'
5973 IE_DESC = False # Do not list
5974 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5975
5976 def _real_extract(self, url):
5977 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5978 return self.url_result(url, 'Generic')
5979
5980
5981class YoutubeTruncatedIDIE(InfoExtractor):
5982 IE_NAME = 'youtube:truncated_id'
5983 IE_DESC = False # Do not list
5984 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5985
5986 _TESTS = [{
5987 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5988 'only_matching': True,
5989 }]
5990
5991 def _real_extract(self, url):
5992 video_id = self._match_id(url)
5993 raise ExtractorError(
5994 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5995 expected=True)