]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[ThumbnailsConvertor] Allow conditional conversion
[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*=\s*({.+})\s*;'
401 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+})\s*;'
402 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
403
404 def _get_default_ytcfg(self, client='web'):
405 return copy.deepcopy(INNERTUBE_CLIENTS[client])
406
407 def _get_innertube_host(self, client='web'):
408 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
409
410 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
411 # try_get but with fallback to default ytcfg client values when present
412 _func = lambda y: try_get(y, getter, expected_type)
413 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
414
415 def _extract_client_name(self, ytcfg, default_client='web'):
416 return self._ytcfg_get_safe(
417 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
418 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
419
420 def _extract_client_version(self, ytcfg, default_client='web'):
421 return self._ytcfg_get_safe(
422 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
423 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
424
425 def _extract_api_key(self, ytcfg=None, default_client='web'):
426 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
427
428 def _extract_context(self, ytcfg=None, default_client='web'):
429 context = get_first(
430 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
431 # Enforce language and tz for extraction
432 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
433 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
434 return context
435
436 _SAPISID = None
437
438 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
439 time_now = round(time.time())
440 if self._SAPISID is None:
441 yt_cookies = self._get_cookies('https://www.youtube.com')
442 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
443 # See: https://github.com/yt-dlp/yt-dlp/issues/393
444 sapisid_cookie = dict_get(
445 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
446 if sapisid_cookie and sapisid_cookie.value:
447 self._SAPISID = sapisid_cookie.value
448 self.write_debug('Extracted SAPISID cookie')
449 # SAPISID cookie is required if not already present
450 if not yt_cookies.get('SAPISID'):
451 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
452 self._set_cookie(
453 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
454 else:
455 self._SAPISID = False
456 if not self._SAPISID:
457 return None
458 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
459 sapisidhash = hashlib.sha1(
460 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
461 return f'SAPISIDHASH {time_now}_{sapisidhash}'
462
463 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
464 note='Downloading API JSON', errnote='Unable to download API page',
465 context=None, api_key=None, api_hostname=None, default_client='web'):
466
467 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
468 data.update(query)
469 real_headers = self.generate_api_headers(default_client=default_client)
470 real_headers.update({'content-type': 'application/json'})
471 if headers:
472 real_headers.update(headers)
473 return self._download_json(
474 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
475 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
476 data=json.dumps(data).encode('utf8'), headers=real_headers,
477 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
478
479 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
480 data = self._search_regex(
481 (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
482 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
483 if data:
484 return self._parse_json(data, item_id, fatal=fatal)
485
486 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
487 return self._parse_json(self._search_regex(
488 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
489 regex), webpage, name, default='{}'), video_id, fatal=False, lenient=True)
490
491 @staticmethod
492 def _extract_session_index(*data):
493 """
494 Index of current account in account list.
495 See: https://github.com/yt-dlp/yt-dlp/pull/519
496 """
497 for ytcfg in data:
498 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
499 if session_index is not None:
500 return session_index
501
502 # Deprecated?
503 def _extract_identity_token(self, ytcfg=None, webpage=None):
504 if ytcfg:
505 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
506 if token:
507 return token
508 if webpage:
509 return self._search_regex(
510 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
511 'identity token', default=None, fatal=False)
512
513 @staticmethod
514 def _extract_account_syncid(*args):
515 """
516 Extract syncId required to download private playlists of secondary channels
517 @params response and/or ytcfg
518 """
519 for data in args:
520 # ytcfg includes channel_syncid if on secondary channel
521 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
522 if delegated_sid:
523 return delegated_sid
524 sync_ids = (try_get(
525 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
526 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
527 if len(sync_ids) >= 2 and sync_ids[1]:
528 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
529 # and just "user_syncid||" for primary channel. We only want the channel_syncid
530 return sync_ids[0]
531
532 @staticmethod
533 def _extract_visitor_data(*args):
534 """
535 Extracts visitorData from an API response or ytcfg
536 Appears to be used to track session state
537 """
538 return get_first(
539 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
540 expected_type=str)
541
542 @functools.cached_property
543 def is_authenticated(self):
544 return bool(self._generate_sapisidhash_header())
545
546 def extract_ytcfg(self, video_id, webpage):
547 if not webpage:
548 return {}
549 return self._parse_json(
550 self._search_regex(
551 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
552 default='{}'), video_id, fatal=False) or {}
553
554 def generate_api_headers(
555 self, *, ytcfg=None, account_syncid=None, session_index=None,
556 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
557
558 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
559 headers = {
560 'X-YouTube-Client-Name': compat_str(
561 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
562 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
563 'Origin': origin,
564 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
565 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
566 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
567 }
568 if session_index is None:
569 session_index = self._extract_session_index(ytcfg)
570 if account_syncid or session_index is not None:
571 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
572
573 auth = self._generate_sapisidhash_header(origin)
574 if auth is not None:
575 headers['Authorization'] = auth
576 headers['X-Origin'] = origin
577 return {h: v for h, v in headers.items() if v is not None}
578
579 def _download_ytcfg(self, client, video_id):
580 url = {
581 'web': 'https://www.youtube.com',
582 'web_music': 'https://music.youtube.com',
583 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
584 }.get(client)
585 if not url:
586 return {}
587 webpage = self._download_webpage(
588 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
589 return self.extract_ytcfg(video_id, webpage) or {}
590
591 @staticmethod
592 def _build_api_continuation_query(continuation, ctp=None):
593 query = {
594 'continuation': continuation
595 }
596 # TODO: Inconsistency with clickTrackingParams.
597 # Currently we have a fixed ctp contained within context (from ytcfg)
598 # and a ctp in root query for continuation.
599 if ctp:
600 query['clickTracking'] = {'clickTrackingParams': ctp}
601 return query
602
603 @classmethod
604 def _extract_next_continuation_data(cls, renderer):
605 next_continuation = try_get(
606 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
607 lambda x: x['continuation']['reloadContinuationData']), dict)
608 if not next_continuation:
609 return
610 continuation = next_continuation.get('continuation')
611 if not continuation:
612 return
613 ctp = next_continuation.get('clickTrackingParams')
614 return cls._build_api_continuation_query(continuation, ctp)
615
616 @classmethod
617 def _extract_continuation_ep_data(cls, continuation_ep: dict):
618 if isinstance(continuation_ep, dict):
619 continuation = try_get(
620 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
621 if not continuation:
622 return
623 ctp = continuation_ep.get('clickTrackingParams')
624 return cls._build_api_continuation_query(continuation, ctp)
625
626 @classmethod
627 def _extract_continuation(cls, renderer):
628 next_continuation = cls._extract_next_continuation_data(renderer)
629 if next_continuation:
630 return next_continuation
631
632 contents = []
633 for key in ('contents', 'items'):
634 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
635
636 for content in contents:
637 if not isinstance(content, dict):
638 continue
639 continuation_ep = try_get(
640 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
641 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
642 dict)
643 continuation = cls._extract_continuation_ep_data(continuation_ep)
644 if continuation:
645 return continuation
646
647 @classmethod
648 def _extract_alerts(cls, data):
649 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
650 if not isinstance(alert_dict, dict):
651 continue
652 for alert in alert_dict.values():
653 alert_type = alert.get('type')
654 if not alert_type:
655 continue
656 message = cls._get_text(alert, 'text')
657 if message:
658 yield alert_type, message
659
660 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
661 errors = []
662 warnings = []
663 for alert_type, alert_message in alerts:
664 if alert_type.lower() == 'error' and fatal:
665 errors.append([alert_type, alert_message])
666 else:
667 warnings.append([alert_type, alert_message])
668
669 for alert_type, alert_message in (warnings + errors[:-1]):
670 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
671 if errors:
672 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
673
674 def _extract_and_report_alerts(self, data, *args, **kwargs):
675 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
676
677 def _extract_badges(self, renderer: dict):
678 badges = set()
679 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
680 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
681 if label:
682 badges.add(label.lower())
683 return badges
684
685 @staticmethod
686 def _get_text(data, *path_list, max_runs=None):
687 for path in path_list or [None]:
688 if path is None:
689 obj = [data]
690 else:
691 obj = traverse_obj(data, path, default=[])
692 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
693 obj = [obj]
694 for item in obj:
695 text = try_get(item, lambda x: x['simpleText'], compat_str)
696 if text:
697 return text
698 runs = try_get(item, lambda x: x['runs'], list) or []
699 if not runs and isinstance(item, list):
700 runs = item
701
702 runs = runs[:min(len(runs), max_runs or len(runs))]
703 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
704 if text:
705 return text
706
707 def _get_count(self, data, *path_list):
708 count_text = self._get_text(data, *path_list) or ''
709 count = parse_count(count_text)
710 if count is None:
711 count = str_to_int(
712 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
713 return count
714
715 @staticmethod
716 def _extract_thumbnails(data, *path_list):
717 """
718 Extract thumbnails from thumbnails dict
719 @param path_list: path list to level that contains 'thumbnails' key
720 """
721 thumbnails = []
722 for path in path_list or [()]:
723 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
724 thumbnail_url = url_or_none(thumbnail.get('url'))
725 if not thumbnail_url:
726 continue
727 # Sometimes youtube gives a wrong thumbnail URL. See:
728 # https://github.com/yt-dlp/yt-dlp/issues/233
729 # https://github.com/ytdl-org/youtube-dl/issues/28023
730 if 'maxresdefault' in thumbnail_url:
731 thumbnail_url = thumbnail_url.split('?')[0]
732 thumbnails.append({
733 'url': thumbnail_url,
734 'height': int_or_none(thumbnail.get('height')),
735 'width': int_or_none(thumbnail.get('width')),
736 })
737 return thumbnails
738
739 @staticmethod
740 def extract_relative_time(relative_time_text):
741 """
742 Extracts a relative time from string and converts to dt object
743 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
744 """
745 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)
746 if mobj:
747 start = mobj.group('start')
748 if start:
749 return datetime_from_str(start)
750 try:
751 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
752 except ValueError:
753 return None
754
755 def _extract_time_text(self, renderer, *path_list):
756 """@returns (timestamp, time_text)"""
757 text = self._get_text(renderer, *path_list) or ''
758 dt = self.extract_relative_time(text)
759 timestamp = None
760 if isinstance(dt, datetime.datetime):
761 timestamp = calendar.timegm(dt.timetuple())
762
763 if timestamp is None:
764 timestamp = (
765 unified_timestamp(text) or unified_timestamp(
766 self._search_regex(
767 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
768 text.lower(), 'time text', default=None)))
769
770 if text and timestamp is None:
771 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
772 return timestamp, text
773
774 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
775 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
776 default_client='web'):
777 response = None
778 last_error = None
779 count = -1
780 retries = self.get_param('extractor_retries', 3)
781 if check_get_keys is None:
782 check_get_keys = []
783 while count < retries:
784 count += 1
785 if last_error:
786 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
787 try:
788 response = self._call_api(
789 ep=ep, fatal=True, headers=headers,
790 video_id=item_id, query=query,
791 context=self._extract_context(ytcfg, default_client),
792 api_key=self._extract_api_key(ytcfg, default_client),
793 api_hostname=api_hostname, default_client=default_client,
794 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
795 except ExtractorError as e:
796 if isinstance(e.cause, network_exceptions):
797 if isinstance(e.cause, compat_HTTPError):
798 first_bytes = e.cause.read(512)
799 if not is_html(first_bytes):
800 yt_error = try_get(
801 self._parse_json(
802 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
803 lambda x: x['error']['message'], compat_str)
804 if yt_error:
805 self._report_alerts([('ERROR', yt_error)], fatal=False)
806 # Downloading page may result in intermittent 5xx HTTP error
807 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
808 # We also want to catch all other network exceptions since errors in later pages can be troublesome
809 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
810 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
811 last_error = error_to_compat_str(e.cause or e.msg)
812 if count < retries:
813 continue
814 if fatal:
815 raise
816 else:
817 self.report_warning(error_to_compat_str(e))
818 return
819
820 else:
821 try:
822 self._extract_and_report_alerts(response, only_once=True)
823 except ExtractorError as e:
824 # YouTube servers may return errors we want to retry on in a 200 OK response
825 # See: https://github.com/yt-dlp/yt-dlp/issues/839
826 if 'unknown error' in e.msg.lower():
827 last_error = e.msg
828 continue
829 if fatal:
830 raise
831 self.report_warning(error_to_compat_str(e))
832 return
833 if not check_get_keys or dict_get(response, check_get_keys):
834 break
835 # Youtube sometimes sends incomplete data
836 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
837 last_error = 'Incomplete data received'
838 if count >= retries:
839 if fatal:
840 raise ExtractorError(last_error)
841 else:
842 self.report_warning(last_error)
843 return
844 return response
845
846 @staticmethod
847 def is_music_url(url):
848 return re.match(r'https?://music\.youtube\.com/', url) is not None
849
850 def _extract_video(self, renderer):
851 video_id = renderer.get('videoId')
852 title = self._get_text(renderer, 'title')
853 description = self._get_text(renderer, 'descriptionSnippet')
854 duration = parse_duration(self._get_text(
855 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
856 if duration is None:
857 duration = parse_duration(self._search_regex(
858 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
859 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
860 video_id, default=None, group='duration'))
861
862 view_count = self._get_count(renderer, 'viewCountText')
863
864 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
865 channel_id = traverse_obj(
866 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
867 expected_type=str, get_all=False)
868 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
869 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
870 overlay_style = traverse_obj(
871 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
872 get_all=False, expected_type=str)
873 badges = self._extract_badges(renderer)
874 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
875 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
876 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
877 expected_type=str)) or ''
878 url = f'https://www.youtube.com/watch?v={video_id}'
879 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
880 url = f'https://www.youtube.com/shorts/{video_id}'
881
882 return {
883 '_type': 'url',
884 'ie_key': YoutubeIE.ie_key(),
885 'id': video_id,
886 'url': url,
887 'title': title,
888 'description': description,
889 'duration': duration,
890 'view_count': view_count,
891 'uploader': uploader,
892 'channel_id': channel_id,
893 'thumbnails': thumbnails,
894 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
895 if self._configuration_arg('approximate_date', ie_key='youtubetab')
896 else None),
897 'live_status': ('is_upcoming' if scheduled_timestamp is not None
898 else 'was_live' if 'streamed' in time_text.lower()
899 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
900 else None),
901 'release_timestamp': scheduled_timestamp,
902 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
903 }
904
905
906class YoutubeIE(YoutubeBaseInfoExtractor):
907 IE_DESC = 'YouTube'
908 _VALID_URL = r"""(?x)^
909 (
910 (?:https?://|//) # http(s):// or protocol-independent URL
911 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
912 (?:www\.)?deturl\.com/www\.youtube\.com|
913 (?:www\.)?pwnyoutube\.com|
914 (?:www\.)?hooktube\.com|
915 (?:www\.)?yourepeat\.com|
916 tube\.majestyc\.net|
917 %(invidious)s|
918 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
919 (?:.*?\#/)? # handle anchor (#/) redirect urls
920 (?: # the various things that can precede the ID:
921 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
922 |(?: # or the v= param in all its forms
923 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
924 (?:\?|\#!?) # the params delimiter ? or # or #!
925 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
926 v=
927 )
928 ))
929 |(?:
930 youtu\.be| # just youtu.be/xxxx
931 vid\.plus| # or vid.plus/xxxx
932 zwearz\.com/watch| # or zwearz.com/watch/xxxx
933 %(invidious)s
934 )/
935 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
936 )
937 )? # all until now is optional -> you can pass the naked ID
938 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
939 (?(1).+)? # if we found the ID, everything can follow
940 (?:\#|$)""" % {
941 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
942 }
943 _PLAYER_INFO_RE = (
944 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
945 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$',
946 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
947 )
948 _formats = {
949 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
950 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
951 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
952 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
953 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
954 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
955 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
956 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
957 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
958 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
959 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
960 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
961 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
962 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
963 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
964 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
965 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
966 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
967
968
969 # 3D videos
970 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
971 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
972 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
973 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
974 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
975 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
976 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
977
978 # Apple HTTP Live Streaming
979 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
980 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
981 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
982 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
983 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
984 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
985 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
986 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
987
988 # DASH mp4 video
989 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
990 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
991 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
992 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
993 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
994 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
995 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
996 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
997 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
998 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
999 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1000 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1001
1002 # Dash mp4 audio
1003 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1004 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1005 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1006 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1007 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1008 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1009 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1010
1011 # Dash webm
1012 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1013 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1014 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1015 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1016 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1017 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1018 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1019 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1020 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1021 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1022 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1023 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1024 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1025 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1026 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1027 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1028 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1029 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1030 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1031 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1032 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1033 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1034
1035 # Dash webm audio
1036 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1037 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1038
1039 # Dash webm audio with opus inside
1040 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1041 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1042 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1043
1044 # RTMP (unnamed)
1045 '_rtmp': {'protocol': 'rtmp'},
1046
1047 # av01 video only formats sometimes served with "unknown" codecs
1048 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1049 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1050 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1051 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1052 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1053 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1054 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1055 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1056 }
1057 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1058
1059 _GEO_BYPASS = False
1060
1061 IE_NAME = 'youtube'
1062 _TESTS = [
1063 {
1064 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1065 'info_dict': {
1066 'id': 'BaW_jenozKc',
1067 'ext': 'mp4',
1068 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1069 'uploader': 'Philipp Hagemeister',
1070 'uploader_id': 'phihag',
1071 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1072 'channel': 'Philipp Hagemeister',
1073 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1074 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1075 'upload_date': '20121002',
1076 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1077 'categories': ['Science & Technology'],
1078 'tags': ['youtube-dl'],
1079 'duration': 10,
1080 'view_count': int,
1081 'like_count': int,
1082 'availability': 'public',
1083 'playable_in_embed': True,
1084 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1085 'live_status': 'not_live',
1086 'age_limit': 0,
1087 'start_time': 1,
1088 'end_time': 9,
1089 'channel_follower_count': int
1090 }
1091 },
1092 {
1093 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1094 'note': 'Embed-only video (#1746)',
1095 'info_dict': {
1096 'id': 'yZIXLfi8CZQ',
1097 'ext': 'mp4',
1098 'upload_date': '20120608',
1099 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1100 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1101 'uploader': 'SET India',
1102 'uploader_id': 'setindia',
1103 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1104 'age_limit': 18,
1105 },
1106 'skip': 'Private video',
1107 },
1108 {
1109 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1110 'note': 'Use the first video ID in the URL',
1111 'info_dict': {
1112 'id': 'BaW_jenozKc',
1113 'ext': 'mp4',
1114 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1115 'uploader': 'Philipp Hagemeister',
1116 'uploader_id': 'phihag',
1117 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1118 'channel': 'Philipp Hagemeister',
1119 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1120 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1121 'upload_date': '20121002',
1122 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1123 'categories': ['Science & Technology'],
1124 'tags': ['youtube-dl'],
1125 'duration': 10,
1126 'view_count': int,
1127 'like_count': int,
1128 'availability': 'public',
1129 'playable_in_embed': True,
1130 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1131 'live_status': 'not_live',
1132 'age_limit': 0,
1133 'channel_follower_count': int
1134 },
1135 'params': {
1136 'skip_download': True,
1137 },
1138 },
1139 {
1140 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1141 'note': '256k DASH audio (format 141) via DASH manifest',
1142 'info_dict': {
1143 'id': 'a9LDPn-MO4I',
1144 'ext': 'm4a',
1145 'upload_date': '20121002',
1146 'uploader_id': '8KVIDEO',
1147 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1148 'description': '',
1149 'uploader': '8KVIDEO',
1150 'title': 'UHDTV TEST 8K VIDEO.mp4'
1151 },
1152 'params': {
1153 'youtube_include_dash_manifest': True,
1154 'format': '141',
1155 },
1156 'skip': 'format 141 not served anymore',
1157 },
1158 # DASH manifest with encrypted signature
1159 {
1160 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1161 'info_dict': {
1162 'id': 'IB3lcPjvWLA',
1163 'ext': 'm4a',
1164 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1165 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1166 'duration': 244,
1167 'uploader': 'AfrojackVEVO',
1168 'uploader_id': 'AfrojackVEVO',
1169 'upload_date': '20131011',
1170 'abr': 129.495,
1171 'like_count': int,
1172 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1173 'playable_in_embed': True,
1174 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1175 'view_count': int,
1176 'track': 'The Spark',
1177 'live_status': 'not_live',
1178 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1179 'channel': 'Afrojack',
1180 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1181 'tags': 'count:19',
1182 'availability': 'public',
1183 'categories': ['Music'],
1184 'age_limit': 0,
1185 'alt_title': 'The Spark',
1186 'channel_follower_count': int
1187 },
1188 'params': {
1189 'youtube_include_dash_manifest': True,
1190 'format': '141/bestaudio[ext=m4a]',
1191 },
1192 },
1193 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1194 {
1195 'note': 'Embed allowed age-gate video',
1196 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1197 'info_dict': {
1198 'id': 'HtVdAasjOgU',
1199 'ext': 'mp4',
1200 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1201 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1202 'duration': 142,
1203 'uploader': 'The Witcher',
1204 'uploader_id': 'WitcherGame',
1205 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1206 'upload_date': '20140605',
1207 'age_limit': 18,
1208 'categories': ['Gaming'],
1209 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1210 'availability': 'needs_auth',
1211 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1212 'like_count': int,
1213 'channel': 'The Witcher',
1214 'live_status': 'not_live',
1215 'tags': 'count:17',
1216 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1217 'playable_in_embed': True,
1218 'view_count': int,
1219 'channel_follower_count': int
1220 },
1221 },
1222 {
1223 'note': 'Age-gate video with embed allowed in public site',
1224 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1225 'info_dict': {
1226 'id': 'HsUATh_Nc2U',
1227 'ext': 'mp4',
1228 'title': 'Godzilla 2 (Official Video)',
1229 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1230 'upload_date': '20200408',
1231 'uploader_id': 'FlyingKitty900',
1232 'uploader': 'FlyingKitty',
1233 'age_limit': 18,
1234 'availability': 'needs_auth',
1235 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1236 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1237 'channel': 'FlyingKitty',
1238 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1239 'view_count': int,
1240 'categories': ['Entertainment'],
1241 'live_status': 'not_live',
1242 'tags': ['Flyingkitty', 'godzilla 2'],
1243 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1244 'like_count': int,
1245 'duration': 177,
1246 'playable_in_embed': True,
1247 'channel_follower_count': int
1248 },
1249 },
1250 {
1251 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1252 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1253 'info_dict': {
1254 'id': 'Tq92D6wQ1mg',
1255 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1256 'ext': 'mp4',
1257 'upload_date': '20191228',
1258 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1259 'uploader': 'Projekt Melody',
1260 'description': 'md5:17eccca93a786d51bc67646756894066',
1261 'age_limit': 18,
1262 'like_count': int,
1263 'availability': 'needs_auth',
1264 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1265 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1266 'view_count': int,
1267 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1268 'channel': 'Projekt Melody',
1269 'live_status': 'not_live',
1270 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1271 'playable_in_embed': True,
1272 'categories': ['Entertainment'],
1273 'duration': 106,
1274 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1275 'channel_follower_count': int
1276 },
1277 },
1278 {
1279 'note': 'Non-Agegated non-embeddable video',
1280 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1281 'info_dict': {
1282 'id': 'MeJVWBSsPAY',
1283 'ext': 'mp4',
1284 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1285 'uploader': 'Herr Lurik',
1286 'uploader_id': 'st3in234',
1287 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1288 'upload_date': '20130730',
1289 'track': 'Such mich find mich',
1290 'age_limit': 0,
1291 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1292 'like_count': int,
1293 'playable_in_embed': False,
1294 'creator': 'OOMPH!',
1295 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1296 'view_count': int,
1297 'alt_title': 'Such mich find mich',
1298 'duration': 210,
1299 'channel': 'Herr Lurik',
1300 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1301 'categories': ['Music'],
1302 'availability': 'public',
1303 'uploader_url': 'http://www.youtube.com/user/st3in234',
1304 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1305 'live_status': 'not_live',
1306 'artist': 'OOMPH!',
1307 'channel_follower_count': int
1308 },
1309 },
1310 {
1311 'note': 'Non-bypassable age-gated video',
1312 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1313 'only_matching': True,
1314 },
1315 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1316 # YouTube Red ad is not captured for creator
1317 {
1318 'url': '__2ABJjxzNo',
1319 'info_dict': {
1320 'id': '__2ABJjxzNo',
1321 'ext': 'mp4',
1322 'duration': 266,
1323 'upload_date': '20100430',
1324 'uploader_id': 'deadmau5',
1325 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1326 'creator': 'deadmau5',
1327 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1328 'uploader': 'deadmau5',
1329 'title': 'Deadmau5 - Some Chords (HD)',
1330 'alt_title': 'Some Chords',
1331 'availability': 'public',
1332 'tags': 'count:14',
1333 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1334 'view_count': int,
1335 'live_status': 'not_live',
1336 'channel': 'deadmau5',
1337 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1338 'like_count': int,
1339 'track': 'Some Chords',
1340 'artist': 'deadmau5',
1341 'playable_in_embed': True,
1342 'age_limit': 0,
1343 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1344 'categories': ['Music'],
1345 'album': 'Some Chords',
1346 'channel_follower_count': int
1347 },
1348 'expected_warnings': [
1349 'DASH manifest missing',
1350 ]
1351 },
1352 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1353 {
1354 'url': 'lqQg6PlCWgI',
1355 'info_dict': {
1356 'id': 'lqQg6PlCWgI',
1357 'ext': 'mp4',
1358 'duration': 6085,
1359 'upload_date': '20150827',
1360 'uploader_id': 'olympic',
1361 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1362 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1363 'uploader': 'Olympics',
1364 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1365 'like_count': int,
1366 'release_timestamp': 1343767800,
1367 'playable_in_embed': True,
1368 'categories': ['Sports'],
1369 'release_date': '20120731',
1370 'channel': 'Olympics',
1371 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1372 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1373 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1374 'age_limit': 0,
1375 'availability': 'public',
1376 'live_status': 'was_live',
1377 'view_count': int,
1378 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1379 'channel_follower_count': int
1380 },
1381 'params': {
1382 'skip_download': 'requires avconv',
1383 }
1384 },
1385 # Non-square pixels
1386 {
1387 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1388 'info_dict': {
1389 'id': '_b-2C3KPAM0',
1390 'ext': 'mp4',
1391 'stretched_ratio': 16 / 9.,
1392 'duration': 85,
1393 'upload_date': '20110310',
1394 'uploader_id': 'AllenMeow',
1395 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1396 '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',
1397 'uploader': '孫ᄋᄅ',
1398 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1399 'playable_in_embed': True,
1400 'channel': '孫ᄋᄅ',
1401 'age_limit': 0,
1402 'tags': 'count:11',
1403 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1404 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1405 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1406 'view_count': int,
1407 'categories': ['People & Blogs'],
1408 'like_count': int,
1409 'live_status': 'not_live',
1410 'availability': 'unlisted',
1411 'channel_follower_count': int
1412 },
1413 },
1414 # url_encoded_fmt_stream_map is empty string
1415 {
1416 'url': 'qEJwOuvDf7I',
1417 'info_dict': {
1418 'id': 'qEJwOuvDf7I',
1419 'ext': 'webm',
1420 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1421 'description': '',
1422 'upload_date': '20150404',
1423 'uploader_id': 'spbelect',
1424 'uploader': 'Наблюдатели Петербурга',
1425 },
1426 'params': {
1427 'skip_download': 'requires avconv',
1428 },
1429 'skip': 'This live event has ended.',
1430 },
1431 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1432 {
1433 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1434 'info_dict': {
1435 'id': 'FIl7x6_3R5Y',
1436 'ext': 'webm',
1437 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1438 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1439 'duration': 220,
1440 'upload_date': '20150625',
1441 'uploader_id': 'dorappi2000',
1442 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1443 'uploader': 'dorappi2000',
1444 'formats': 'mincount:31',
1445 },
1446 'skip': 'not actual anymore',
1447 },
1448 # DASH manifest with segment_list
1449 {
1450 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1451 'md5': '8ce563a1d667b599d21064e982ab9e31',
1452 'info_dict': {
1453 'id': 'CsmdDsKjzN8',
1454 'ext': 'mp4',
1455 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1456 'uploader': 'Airtek',
1457 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1458 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1459 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1460 },
1461 'params': {
1462 'youtube_include_dash_manifest': True,
1463 'format': '135', # bestvideo
1464 },
1465 'skip': 'This live event has ended.',
1466 },
1467 {
1468 # Multifeed videos (multiple cameras), URL is for Main Camera
1469 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1470 'info_dict': {
1471 'id': 'jvGDaLqkpTg',
1472 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1473 'description': 'md5:e03b909557865076822aa169218d6a5d',
1474 },
1475 'playlist': [{
1476 'info_dict': {
1477 'id': 'jvGDaLqkpTg',
1478 'ext': 'mp4',
1479 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1480 'description': 'md5:e03b909557865076822aa169218d6a5d',
1481 'duration': 10643,
1482 'upload_date': '20161111',
1483 'uploader': 'Team PGP',
1484 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1485 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1486 },
1487 }, {
1488 'info_dict': {
1489 'id': '3AKt1R1aDnw',
1490 'ext': 'mp4',
1491 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1492 'description': 'md5:e03b909557865076822aa169218d6a5d',
1493 'duration': 10991,
1494 'upload_date': '20161111',
1495 'uploader': 'Team PGP',
1496 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1497 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1498 },
1499 }, {
1500 'info_dict': {
1501 'id': 'RtAMM00gpVc',
1502 'ext': 'mp4',
1503 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1504 'description': 'md5:e03b909557865076822aa169218d6a5d',
1505 'duration': 10995,
1506 'upload_date': '20161111',
1507 'uploader': 'Team PGP',
1508 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1509 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1510 },
1511 }, {
1512 'info_dict': {
1513 'id': '6N2fdlP3C5U',
1514 'ext': 'mp4',
1515 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1516 'description': 'md5:e03b909557865076822aa169218d6a5d',
1517 'duration': 10990,
1518 'upload_date': '20161111',
1519 'uploader': 'Team PGP',
1520 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1521 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1522 },
1523 }],
1524 'params': {
1525 'skip_download': True,
1526 },
1527 'skip': 'Not multifeed anymore',
1528 },
1529 {
1530 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1531 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1532 'info_dict': {
1533 'id': 'gVfLd0zydlo',
1534 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1535 },
1536 'playlist_count': 2,
1537 'skip': 'Not multifeed anymore',
1538 },
1539 {
1540 'url': 'https://vid.plus/FlRa-iH7PGw',
1541 'only_matching': True,
1542 },
1543 {
1544 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1545 'only_matching': True,
1546 },
1547 {
1548 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1549 # Also tests cut-off URL expansion in video description (see
1550 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1551 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1552 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1553 'info_dict': {
1554 'id': 'lsguqyKfVQg',
1555 'ext': 'mp4',
1556 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1557 'alt_title': 'Dark Walk',
1558 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1559 'duration': 133,
1560 'upload_date': '20151119',
1561 'uploader_id': 'IronSoulElf',
1562 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1563 'uploader': 'IronSoulElf',
1564 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1565 'track': 'Dark Walk',
1566 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1567 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1568 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1569 'categories': ['Film & Animation'],
1570 'view_count': int,
1571 'live_status': 'not_live',
1572 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1573 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1574 'tags': 'count:13',
1575 'availability': 'public',
1576 'channel': 'IronSoulElf',
1577 'playable_in_embed': True,
1578 'like_count': int,
1579 'age_limit': 0,
1580 'channel_follower_count': int
1581 },
1582 'params': {
1583 'skip_download': True,
1584 },
1585 },
1586 {
1587 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1588 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1589 'only_matching': True,
1590 },
1591 {
1592 # Video with yt:stretch=17:0
1593 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1594 'info_dict': {
1595 'id': 'Q39EVAstoRM',
1596 'ext': 'mp4',
1597 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1598 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1599 'upload_date': '20151107',
1600 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1601 'uploader': 'CH GAMER DROID',
1602 },
1603 'params': {
1604 'skip_download': True,
1605 },
1606 'skip': 'This video does not exist.',
1607 },
1608 {
1609 # Video with incomplete 'yt:stretch=16:'
1610 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1611 'only_matching': True,
1612 },
1613 {
1614 # Video licensed under Creative Commons
1615 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1616 'info_dict': {
1617 'id': 'M4gD1WSo5mA',
1618 'ext': 'mp4',
1619 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1620 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1621 'duration': 721,
1622 'upload_date': '20150128',
1623 'uploader_id': 'BerkmanCenter',
1624 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1625 'uploader': 'The Berkman Klein Center for Internet & Society',
1626 'license': 'Creative Commons Attribution license (reuse allowed)',
1627 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1628 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1629 'like_count': int,
1630 'age_limit': 0,
1631 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1632 'channel': 'The Berkman Klein Center for Internet & Society',
1633 'availability': 'public',
1634 'view_count': int,
1635 'categories': ['Education'],
1636 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1637 'live_status': 'not_live',
1638 'playable_in_embed': True,
1639 'channel_follower_count': int
1640 },
1641 'params': {
1642 'skip_download': True,
1643 },
1644 },
1645 {
1646 # Channel-like uploader_url
1647 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1648 'info_dict': {
1649 'id': 'eQcmzGIKrzg',
1650 'ext': 'mp4',
1651 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1652 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1653 'duration': 4060,
1654 'upload_date': '20151120',
1655 'uploader': 'Bernie Sanders',
1656 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1657 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1658 'license': 'Creative Commons Attribution license (reuse allowed)',
1659 'playable_in_embed': True,
1660 'tags': 'count:12',
1661 'like_count': int,
1662 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1663 'age_limit': 0,
1664 'availability': 'public',
1665 'categories': ['News & Politics'],
1666 'channel': 'Bernie Sanders',
1667 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1668 'view_count': int,
1669 'live_status': 'not_live',
1670 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1671 'channel_follower_count': int
1672 },
1673 'params': {
1674 'skip_download': True,
1675 },
1676 },
1677 {
1678 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1679 'only_matching': True,
1680 },
1681 {
1682 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1683 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1684 'only_matching': True,
1685 },
1686 {
1687 # Rental video preview
1688 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1689 'info_dict': {
1690 'id': 'uGpuVWrhIzE',
1691 'ext': 'mp4',
1692 'title': 'Piku - Trailer',
1693 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1694 'upload_date': '20150811',
1695 'uploader': 'FlixMatrix',
1696 'uploader_id': 'FlixMatrixKaravan',
1697 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1698 'license': 'Standard YouTube License',
1699 },
1700 'params': {
1701 'skip_download': True,
1702 },
1703 'skip': 'This video is not available.',
1704 },
1705 {
1706 # YouTube Red video with episode data
1707 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1708 'info_dict': {
1709 'id': 'iqKdEhx-dD4',
1710 'ext': 'mp4',
1711 'title': 'Isolation - Mind Field (Ep 1)',
1712 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1713 'duration': 2085,
1714 'upload_date': '20170118',
1715 'uploader': 'Vsauce',
1716 'uploader_id': 'Vsauce',
1717 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1718 'series': 'Mind Field',
1719 'season_number': 1,
1720 'episode_number': 1,
1721 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1722 'tags': 'count:12',
1723 'view_count': int,
1724 'availability': 'public',
1725 'age_limit': 0,
1726 'channel': 'Vsauce',
1727 'episode': 'Episode 1',
1728 'categories': ['Entertainment'],
1729 'season': 'Season 1',
1730 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1731 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1732 'like_count': int,
1733 'playable_in_embed': True,
1734 'live_status': 'not_live',
1735 'channel_follower_count': int
1736 },
1737 'params': {
1738 'skip_download': True,
1739 },
1740 'expected_warnings': [
1741 'Skipping DASH manifest',
1742 ],
1743 },
1744 {
1745 # The following content has been identified by the YouTube community
1746 # as inappropriate or offensive to some audiences.
1747 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1748 'info_dict': {
1749 'id': '6SJNVb0GnPI',
1750 'ext': 'mp4',
1751 'title': 'Race Differences in Intelligence',
1752 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1753 'duration': 965,
1754 'upload_date': '20140124',
1755 'uploader': 'New Century Foundation',
1756 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1757 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1758 },
1759 'params': {
1760 'skip_download': True,
1761 },
1762 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1763 },
1764 {
1765 # itag 212
1766 'url': '1t24XAntNCY',
1767 'only_matching': True,
1768 },
1769 {
1770 # geo restricted to JP
1771 'url': 'sJL6WA-aGkQ',
1772 'only_matching': True,
1773 },
1774 {
1775 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1776 'only_matching': True,
1777 },
1778 {
1779 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1780 'only_matching': True,
1781 },
1782 {
1783 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1784 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1785 'only_matching': True,
1786 },
1787 {
1788 # DRM protected
1789 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1790 'only_matching': True,
1791 },
1792 {
1793 # Video with unsupported adaptive stream type formats
1794 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1795 'info_dict': {
1796 'id': 'Z4Vy8R84T1U',
1797 'ext': 'mp4',
1798 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1799 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1800 'duration': 433,
1801 'upload_date': '20130923',
1802 'uploader': 'Amelia Putri Harwita',
1803 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1804 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1805 'formats': 'maxcount:10',
1806 },
1807 'params': {
1808 'skip_download': True,
1809 'youtube_include_dash_manifest': False,
1810 },
1811 'skip': 'not actual anymore',
1812 },
1813 {
1814 # Youtube Music Auto-generated description
1815 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1816 'info_dict': {
1817 'id': 'MgNrAu2pzNs',
1818 'ext': 'mp4',
1819 'title': 'Voyeur Girl',
1820 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1821 'upload_date': '20190312',
1822 'uploader': 'Stephen - Topic',
1823 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1824 'artist': 'Stephen',
1825 'track': 'Voyeur Girl',
1826 'album': 'it\'s too much love to know my dear',
1827 'release_date': '20190313',
1828 'release_year': 2019,
1829 'alt_title': 'Voyeur Girl',
1830 'view_count': int,
1831 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1832 'playable_in_embed': True,
1833 'like_count': int,
1834 'categories': ['Music'],
1835 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1836 'channel': 'Stephen',
1837 'availability': 'public',
1838 'creator': 'Stephen',
1839 'duration': 169,
1840 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1841 'age_limit': 0,
1842 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1843 'tags': 'count:11',
1844 'live_status': 'not_live',
1845 'channel_follower_count': int
1846 },
1847 'params': {
1848 'skip_download': True,
1849 },
1850 },
1851 {
1852 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1853 'only_matching': True,
1854 },
1855 {
1856 # invalid -> valid video id redirection
1857 'url': 'DJztXj2GPfl',
1858 'info_dict': {
1859 'id': 'DJztXj2GPfk',
1860 'ext': 'mp4',
1861 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1862 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1863 'upload_date': '20090125',
1864 'uploader': 'Prochorowka',
1865 'uploader_id': 'Prochorowka',
1866 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1867 'artist': 'Panjabi MC',
1868 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1869 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1870 },
1871 'params': {
1872 'skip_download': True,
1873 },
1874 'skip': 'Video unavailable',
1875 },
1876 {
1877 # empty description results in an empty string
1878 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1879 'info_dict': {
1880 'id': 'x41yOUIvK2k',
1881 'ext': 'mp4',
1882 'title': 'IMG 3456',
1883 'description': '',
1884 'upload_date': '20170613',
1885 'uploader_id': 'ElevageOrVert',
1886 'uploader': 'ElevageOrVert',
1887 'view_count': int,
1888 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1889 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1890 'like_count': int,
1891 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1892 'tags': [],
1893 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1894 'availability': 'public',
1895 'age_limit': 0,
1896 'categories': ['Pets & Animals'],
1897 'duration': 7,
1898 'playable_in_embed': True,
1899 'live_status': 'not_live',
1900 'channel': 'ElevageOrVert',
1901 'channel_follower_count': int
1902 },
1903 'params': {
1904 'skip_download': True,
1905 },
1906 },
1907 {
1908 # with '};' inside yt initial data (see [1])
1909 # see [2] for an example with '};' inside ytInitialPlayerResponse
1910 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1911 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1912 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1913 'info_dict': {
1914 'id': 'CHqg6qOn4no',
1915 'ext': 'mp4',
1916 'title': 'Part 77 Sort a list of simple types in c#',
1917 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1918 'upload_date': '20130831',
1919 'uploader_id': 'kudvenkat',
1920 'uploader': 'kudvenkat',
1921 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1922 'like_count': int,
1923 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1924 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1925 'live_status': 'not_live',
1926 'categories': ['Education'],
1927 'availability': 'public',
1928 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1929 'tags': 'count:12',
1930 'playable_in_embed': True,
1931 'age_limit': 0,
1932 'view_count': int,
1933 'duration': 522,
1934 'channel': 'kudvenkat',
1935 'channel_follower_count': int
1936 },
1937 'params': {
1938 'skip_download': True,
1939 },
1940 },
1941 {
1942 # another example of '};' in ytInitialData
1943 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1944 'only_matching': True,
1945 },
1946 {
1947 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1948 'only_matching': True,
1949 },
1950 {
1951 # https://github.com/ytdl-org/youtube-dl/pull/28094
1952 'url': 'OtqTfy26tG0',
1953 'info_dict': {
1954 'id': 'OtqTfy26tG0',
1955 'ext': 'mp4',
1956 'title': 'Burn Out',
1957 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1958 'upload_date': '20141120',
1959 'uploader': 'The Cinematic Orchestra - Topic',
1960 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1961 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1962 'artist': 'The Cinematic Orchestra',
1963 'track': 'Burn Out',
1964 'album': 'Every Day',
1965 'like_count': int,
1966 'live_status': 'not_live',
1967 'alt_title': 'Burn Out',
1968 'duration': 614,
1969 'age_limit': 0,
1970 'view_count': int,
1971 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1972 'creator': 'The Cinematic Orchestra',
1973 'channel': 'The Cinematic Orchestra',
1974 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1975 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1976 'availability': 'public',
1977 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1978 'categories': ['Music'],
1979 'playable_in_embed': True,
1980 'channel_follower_count': int
1981 },
1982 'params': {
1983 'skip_download': True,
1984 },
1985 },
1986 {
1987 # controversial video, only works with bpctr when authenticated with cookies
1988 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1989 'only_matching': True,
1990 },
1991 {
1992 # controversial video, requires bpctr/contentCheckOk
1993 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1994 'info_dict': {
1995 'id': 'SZJvDhaSDnc',
1996 'ext': 'mp4',
1997 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1998 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1999 'uploader': 'CBS Mornings',
2000 'uploader_id': 'CBSThisMorning',
2001 'upload_date': '20140716',
2002 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2003 'duration': 170,
2004 'categories': ['News & Politics'],
2005 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2006 'view_count': int,
2007 'channel': 'CBS Mornings',
2008 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2009 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2010 'age_limit': 18,
2011 'availability': 'needs_auth',
2012 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2013 'like_count': int,
2014 'live_status': 'not_live',
2015 'playable_in_embed': True,
2016 'channel_follower_count': int
2017 }
2018 },
2019 {
2020 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2021 'url': 'cBvYw8_A0vQ',
2022 'info_dict': {
2023 'id': 'cBvYw8_A0vQ',
2024 'ext': 'mp4',
2025 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2026 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2027 'upload_date': '20201120',
2028 'uploader': 'Walk around Japan',
2029 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2030 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2031 'duration': 1456,
2032 'categories': ['Travel & Events'],
2033 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2034 'view_count': int,
2035 'channel': 'Walk around Japan',
2036 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2037 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2038 'age_limit': 0,
2039 'availability': 'public',
2040 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2041 'live_status': 'not_live',
2042 'playable_in_embed': True,
2043 'channel_follower_count': int
2044 },
2045 'params': {
2046 'skip_download': True,
2047 },
2048 }, {
2049 # Has multiple audio streams
2050 'url': 'WaOKSUlf4TM',
2051 'only_matching': True
2052 }, {
2053 # Requires Premium: has format 141 when requested using YTM url
2054 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2055 'only_matching': True
2056 }, {
2057 # multiple subtitles with same lang_code
2058 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2059 'only_matching': True,
2060 }, {
2061 # Force use android client fallback
2062 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2063 'info_dict': {
2064 'id': 'YOelRv7fMxY',
2065 'title': 'DIGGING A SECRET TUNNEL Part 1',
2066 'ext': '3gp',
2067 'upload_date': '20210624',
2068 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2069 'uploader': 'colinfurze',
2070 'uploader_id': 'colinfurze',
2071 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2072 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2073 'duration': 596,
2074 'categories': ['Entertainment'],
2075 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2076 'view_count': int,
2077 'channel': 'colinfurze',
2078 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2079 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2080 'age_limit': 0,
2081 'availability': 'public',
2082 'like_count': int,
2083 'live_status': 'not_live',
2084 'playable_in_embed': True,
2085 'channel_follower_count': int
2086 },
2087 'params': {
2088 'format': '17', # 3gp format available on android
2089 'extractor_args': {'youtube': {'player_client': ['android']}},
2090 },
2091 },
2092 {
2093 # Skip download of additional client configs (remix client config in this case)
2094 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2095 'only_matching': True,
2096 'params': {
2097 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2098 },
2099 }, {
2100 # shorts
2101 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2102 'only_matching': True,
2103 }, {
2104 'note': 'Storyboards',
2105 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2106 'info_dict': {
2107 'id': '5KLPxDtMqe8',
2108 'ext': 'mhtml',
2109 'format_id': 'sb0',
2110 'title': 'Your Brain is Plastic',
2111 'uploader_id': 'scishow',
2112 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2113 'upload_date': '20140324',
2114 'uploader': 'SciShow',
2115 'like_count': int,
2116 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2117 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2118 'view_count': int,
2119 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2120 'playable_in_embed': True,
2121 'tags': 'count:12',
2122 'uploader_url': 'http://www.youtube.com/user/scishow',
2123 'availability': 'public',
2124 'channel': 'SciShow',
2125 'live_status': 'not_live',
2126 'duration': 248,
2127 'categories': ['Education'],
2128 'age_limit': 0,
2129 'channel_follower_count': int
2130 }, 'params': {'format': 'mhtml', 'skip_download': True}
2131 }, {
2132 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2133 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2134 'info_dict': {
2135 'id': '2NUZ8W2llS4',
2136 'ext': 'mp4',
2137 'title': 'The NP that test your phone performance 🙂',
2138 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2139 'uploader': 'Leon Nguyen',
2140 'uploader_id': 'VNSXIII',
2141 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2142 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2143 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2144 'duration': 21,
2145 'view_count': int,
2146 'age_limit': 0,
2147 'categories': ['Gaming'],
2148 'tags': 'count:23',
2149 'playable_in_embed': True,
2150 'live_status': 'not_live',
2151 'upload_date': '20220103',
2152 'like_count': int,
2153 'availability': 'public',
2154 'channel': 'Leon Nguyen',
2155 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2156 'channel_follower_count': int
2157 }
2158 }, {
2159 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2160 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2161 'info_dict': {
2162 'id': 'mzZzzBU6lrM',
2163 'ext': 'mp4',
2164 'title': 'I Met GeorgeNotFound In Real Life...',
2165 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2166 'uploader': 'Quackity',
2167 'uploader_id': 'QuackityHQ',
2168 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2169 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2170 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2171 'duration': 955,
2172 'view_count': int,
2173 'age_limit': 0,
2174 'categories': ['Entertainment'],
2175 'tags': 'count:26',
2176 'playable_in_embed': True,
2177 'live_status': 'not_live',
2178 'release_timestamp': 1641172509,
2179 'release_date': '20220103',
2180 'upload_date': '20220103',
2181 'like_count': int,
2182 'availability': 'public',
2183 'channel': 'Quackity',
2184 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2185 'channel_follower_count': int
2186 }
2187 },
2188 { # continuous livestream. Microformat upload date should be preferred.
2189 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2190 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2191 'info_dict': {
2192 'id': 'kgx4WGK0oNU',
2193 '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}',
2194 'ext': 'mp4',
2195 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2196 'availability': 'public',
2197 'age_limit': 0,
2198 'release_timestamp': 1637975704,
2199 'upload_date': '20210619',
2200 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2201 'live_status': 'is_live',
2202 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2203 'uploader': '阿鲍Abao',
2204 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2205 'channel': 'Abao in Tokyo',
2206 'channel_follower_count': int,
2207 'release_date': '20211127',
2208 'tags': 'count:39',
2209 'categories': ['People & Blogs'],
2210 'like_count': int,
2211 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2212 'view_count': int,
2213 'playable_in_embed': True,
2214 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2215 },
2216 'params': {'skip_download': True}
2217 }, {
2218 # Story. Requires specific player params to work.
2219 # Note: stories get removed after some period of time
2220 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2221 'info_dict': {
2222 'id': 'vv8qTUWmulI',
2223 'ext': 'mp4',
2224 'availability': 'unlisted',
2225 'view_count': int,
2226 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2227 'upload_date': '20220526',
2228 'categories': ['Education'],
2229 'title': 'Story',
2230 'channel': 'IT\'S HISTORY',
2231 'description': '',
2232 'uploader_id': 'BlastfromthePast',
2233 'duration': 12,
2234 'uploader': 'IT\'S HISTORY',
2235 'playable_in_embed': True,
2236 'age_limit': 0,
2237 'live_status': 'not_live',
2238 'tags': [],
2239 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2240 'uploader_url': 'http://www.youtube.com/user/BlastfromthePast',
2241 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2242 }
2243 }, {
2244 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2245 'info_dict': {
2246 'id': 'tjjjtzRLHvA',
2247 'ext': 'mp4',
2248 'title': 'ハッシュタグ無し };if window.ytcsi',
2249 'upload_date': '20220323',
2250 'like_count': int,
2251 'availability': 'unlisted',
2252 'channel': 'nao20010128nao',
2253 'thumbnail': 'https://i.ytimg.com/vi_webp/tjjjtzRLHvA/maxresdefault.webp',
2254 'age_limit': 0,
2255 'uploader': 'nao20010128nao',
2256 'uploader_id': 'nao20010128nao',
2257 'categories': ['Music'],
2258 'view_count': int,
2259 'description': '',
2260 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2261 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2262 'live_status': 'not_live',
2263 'playable_in_embed': True,
2264 'channel_follower_count': int,
2265 'duration': 6,
2266 'tags': [],
2267 'uploader_url': 'http://www.youtube.com/user/nao20010128nao',
2268 }
2269 }
2270 ]
2271
2272 @classmethod
2273 def suitable(cls, url):
2274 from ..utils import parse_qs
2275
2276 qs = parse_qs(url)
2277 if qs.get('list', [None])[0]:
2278 return False
2279 return super().suitable(url)
2280
2281 def __init__(self, *args, **kwargs):
2282 super().__init__(*args, **kwargs)
2283 self._code_cache = {}
2284 self._player_cache = {}
2285
2286 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2287 lock = threading.Lock()
2288
2289 is_live = True
2290 start_time = time.time()
2291 formats = [f for f in formats if f.get('is_from_start')]
2292
2293 def refetch_manifest(format_id, delay):
2294 nonlocal formats, start_time, is_live
2295 if time.time() <= start_time + delay:
2296 return
2297
2298 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2299 video_details = traverse_obj(
2300 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2301 microformats = traverse_obj(
2302 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2303 expected_type=dict, default=[])
2304 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2305 start_time = time.time()
2306
2307 def mpd_feed(format_id, delay):
2308 """
2309 @returns (manifest_url, manifest_stream_number, is_live) or None
2310 """
2311 with lock:
2312 refetch_manifest(format_id, delay)
2313
2314 f = next((f for f in formats if f['format_id'] == format_id), None)
2315 if not f:
2316 if not is_live:
2317 self.to_screen(f'{video_id}: Video is no longer live')
2318 else:
2319 self.report_warning(
2320 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2321 return None
2322 return f['manifest_url'], f['manifest_stream_number'], is_live
2323
2324 for f in formats:
2325 f['is_live'] = True
2326 f['protocol'] = 'http_dash_segments_generator'
2327 f['fragments'] = functools.partial(
2328 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2329
2330 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2331 FETCH_SPAN, MAX_DURATION = 5, 432000
2332
2333 mpd_url, stream_number, is_live = None, None, True
2334
2335 begin_index = 0
2336 download_start_time = ctx.get('start') or time.time()
2337
2338 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2339 if lack_early_segments:
2340 self.report_warning(bug_reports_message(
2341 'Starting download from the last 120 hours of the live stream since '
2342 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2343 lack_early_segments = True
2344
2345 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2346 fragments, fragment_base_url = None, None
2347
2348 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2349 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2350 # Obtain from MPD's maximum seq value
2351 old_mpd_url = mpd_url
2352 last_error = ctx.pop('last_error', None)
2353 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2354 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2355 or (mpd_url, stream_number, False))
2356 if not refresh_sequence:
2357 if expire_fast and not is_live:
2358 return False, last_seq
2359 elif old_mpd_url == mpd_url:
2360 return True, last_seq
2361 try:
2362 fmts, _ = self._extract_mpd_formats_and_subtitles(
2363 mpd_url, None, note=False, errnote=False, fatal=False)
2364 except ExtractorError:
2365 fmts = None
2366 if not fmts:
2367 no_fragment_score += 2
2368 return False, last_seq
2369 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2370 fragments = fmt_info['fragments']
2371 fragment_base_url = fmt_info['fragment_base_url']
2372 assert fragment_base_url
2373
2374 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2375 return True, _last_seq
2376
2377 while is_live:
2378 fetch_time = time.time()
2379 if no_fragment_score > 30:
2380 return
2381 if last_segment_url:
2382 # Obtain from "X-Head-Seqnum" header value from each segment
2383 try:
2384 urlh = self._request_webpage(
2385 last_segment_url, None, note=False, errnote=False, fatal=False)
2386 except ExtractorError:
2387 urlh = None
2388 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2389 if last_seq is None:
2390 no_fragment_score += 2
2391 last_segment_url = None
2392 continue
2393 else:
2394 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2395 no_fragment_score += 2
2396 if not should_continue:
2397 continue
2398
2399 if known_idx > last_seq:
2400 last_segment_url = None
2401 continue
2402
2403 last_seq += 1
2404
2405 if begin_index < 0 and known_idx < 0:
2406 # skip from the start when it's negative value
2407 known_idx = last_seq + begin_index
2408 if lack_early_segments:
2409 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2410 try:
2411 for idx in range(known_idx, last_seq):
2412 # do not update sequence here or you'll get skipped some part of it
2413 should_continue, _ = _extract_sequence_from_mpd(False, False)
2414 if not should_continue:
2415 known_idx = idx - 1
2416 raise ExtractorError('breaking out of outer loop')
2417 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2418 yield {
2419 'url': last_segment_url,
2420 }
2421 if known_idx == last_seq:
2422 no_fragment_score += 5
2423 else:
2424 no_fragment_score = 0
2425 known_idx = last_seq
2426 except ExtractorError:
2427 continue
2428
2429 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2430
2431 def _extract_player_url(self, *ytcfgs, webpage=None):
2432 player_url = traverse_obj(
2433 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2434 get_all=False, expected_type=compat_str)
2435 if not player_url:
2436 return
2437 return urljoin('https://www.youtube.com', player_url)
2438
2439 def _download_player_url(self, video_id, fatal=False):
2440 res = self._download_webpage(
2441 'https://www.youtube.com/iframe_api',
2442 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2443 if res:
2444 player_version = self._search_regex(
2445 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2446 if player_version:
2447 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2448
2449 def _signature_cache_id(self, example_sig):
2450 """ Return a string representation of a signature """
2451 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2452
2453 @classmethod
2454 def _extract_player_info(cls, player_url):
2455 for player_re in cls._PLAYER_INFO_RE:
2456 id_m = re.search(player_re, player_url)
2457 if id_m:
2458 break
2459 else:
2460 raise ExtractorError('Cannot identify player %r' % player_url)
2461 return id_m.group('id')
2462
2463 def _load_player(self, video_id, player_url, fatal=True):
2464 player_id = self._extract_player_info(player_url)
2465 if player_id not in self._code_cache:
2466 code = self._download_webpage(
2467 player_url, video_id, fatal=fatal,
2468 note='Downloading player ' + player_id,
2469 errnote='Download of %s failed' % player_url)
2470 if code:
2471 self._code_cache[player_id] = code
2472 return self._code_cache.get(player_id)
2473
2474 def _extract_signature_function(self, video_id, player_url, example_sig):
2475 player_id = self._extract_player_info(player_url)
2476
2477 # Read from filesystem cache
2478 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2479 assert os.path.basename(func_id) == func_id
2480
2481 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2482 if cache_spec is not None:
2483 return lambda s: ''.join(s[i] for i in cache_spec)
2484
2485 code = self._load_player(video_id, player_url)
2486 if code:
2487 res = self._parse_sig_js(code)
2488
2489 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2490 cache_res = res(test_string)
2491 cache_spec = [ord(c) for c in cache_res]
2492
2493 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2494 return res
2495
2496 def _print_sig_code(self, func, example_sig):
2497 if not self.get_param('youtube_print_sig_code'):
2498 return
2499
2500 def gen_sig_code(idxs):
2501 def _genslice(start, end, step):
2502 starts = '' if start == 0 else str(start)
2503 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2504 steps = '' if step == 1 else (':%d' % step)
2505 return f's[{starts}{ends}{steps}]'
2506
2507 step = None
2508 # Quelch pyflakes warnings - start will be set when step is set
2509 start = '(Never used)'
2510 for i, prev in zip(idxs[1:], idxs[:-1]):
2511 if step is not None:
2512 if i - prev == step:
2513 continue
2514 yield _genslice(start, prev, step)
2515 step = None
2516 continue
2517 if i - prev in [-1, 1]:
2518 step = i - prev
2519 start = prev
2520 continue
2521 else:
2522 yield 's[%d]' % prev
2523 if step is None:
2524 yield 's[%d]' % i
2525 else:
2526 yield _genslice(start, i, step)
2527
2528 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2529 cache_res = func(test_string)
2530 cache_spec = [ord(c) for c in cache_res]
2531 expr_code = ' + '.join(gen_sig_code(cache_spec))
2532 signature_id_tuple = '(%s)' % (
2533 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2534 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2535 ' return %s\n') % (signature_id_tuple, expr_code)
2536 self.to_screen('Extracted signature function:\n' + code)
2537
2538 def _parse_sig_js(self, jscode):
2539 funcname = self._search_regex(
2540 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2541 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2542 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2543 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2544 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}\(a,\d+\)',
2545 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\(\s*""\s*\)',
2546 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2547 # Obsolete patterns
2548 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2549 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2550 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2551 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2552 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2553 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2554 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2555 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2556 jscode, 'Initial JS player signature function name', group='sig')
2557
2558 jsi = JSInterpreter(jscode)
2559 initial_function = jsi.extract_function(funcname)
2560 return lambda s: initial_function([s])
2561
2562 def _decrypt_signature(self, s, video_id, player_url):
2563 """Turn the encrypted s field into a working signature"""
2564 try:
2565 player_id = (player_url, self._signature_cache_id(s))
2566 if player_id not in self._player_cache:
2567 func = self._extract_signature_function(video_id, player_url, s)
2568 self._player_cache[player_id] = func
2569 func = self._player_cache[player_id]
2570 self._print_sig_code(func, s)
2571 return func(s)
2572 except Exception as e:
2573 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2574
2575 def _decrypt_nsig(self, s, video_id, player_url):
2576 """Turn the encrypted n field into a working signature"""
2577 if player_url is None:
2578 raise ExtractorError('Cannot decrypt nsig without player_url')
2579 player_url = urljoin('https://www.youtube.com', player_url)
2580
2581 sig_id = ('nsig_value', s)
2582 if sig_id in self._player_cache:
2583 return self._player_cache[sig_id]
2584
2585 try:
2586 player_id = ('nsig', player_url)
2587 if player_id not in self._player_cache:
2588 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2589 func = self._player_cache[player_id]
2590 self._player_cache[sig_id] = func(s)
2591 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2592 return self._player_cache[sig_id]
2593 except Exception as e:
2594 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2595
2596 def _extract_n_function_name(self, jscode):
2597 nfunc, idx = self._search_regex(
2598 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2599 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2600 if not idx:
2601 return nfunc
2602 return json.loads(js_to_json(self._search_regex(
2603 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2604 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2605
2606 def _extract_n_function(self, video_id, player_url):
2607 player_id = self._extract_player_info(player_url)
2608 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2609
2610 if func_code:
2611 jsi = JSInterpreter(func_code)
2612 else:
2613 jscode = self._load_player(video_id, player_url)
2614 funcname = self._extract_n_function_name(jscode)
2615 jsi = JSInterpreter(jscode)
2616 func_code = jsi.extract_function_code(funcname)
2617 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2618
2619 if self.get_param('youtube_print_sig_code'):
2620 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2621
2622 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2623
2624 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2625 """
2626 Extract signatureTimestamp (sts)
2627 Required to tell API what sig/player version is in use.
2628 """
2629 sts = None
2630 if isinstance(ytcfg, dict):
2631 sts = int_or_none(ytcfg.get('STS'))
2632
2633 if not sts:
2634 # Attempt to extract from player
2635 if player_url is None:
2636 error_msg = 'Cannot extract signature timestamp without player_url.'
2637 if fatal:
2638 raise ExtractorError(error_msg)
2639 self.report_warning(error_msg)
2640 return
2641 code = self._load_player(video_id, player_url, fatal=fatal)
2642 if code:
2643 sts = int_or_none(self._search_regex(
2644 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2645 'JS player signature timestamp', group='sts', fatal=fatal))
2646 return sts
2647
2648 def _mark_watched(self, video_id, player_responses):
2649 playback_url = get_first(
2650 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2651 expected_type=url_or_none)
2652 if not playback_url:
2653 self.report_warning('Unable to mark watched')
2654 return
2655 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2656 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2657
2658 # cpn generation algorithm is reverse engineered from base.js.
2659 # In fact it works even with dummy cpn.
2660 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2661 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2662
2663 qs.update({
2664 'ver': ['2'],
2665 'cpn': [cpn],
2666 })
2667 playback_url = compat_urlparse.urlunparse(
2668 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2669
2670 self._download_webpage(
2671 playback_url, video_id, 'Marking watched',
2672 'Unable to mark watched', fatal=False)
2673
2674 @staticmethod
2675 def _extract_urls(webpage):
2676 # Embedded YouTube player
2677 entries = [
2678 unescapeHTML(mobj.group('url'))
2679 for mobj in re.finditer(r'''(?x)
2680 (?:
2681 <iframe[^>]+?src=|
2682 data-video-url=|
2683 <embed[^>]+?src=|
2684 embedSWF\(?:\s*|
2685 <object[^>]+data=|
2686 new\s+SWFObject\(
2687 )
2688 (["\'])
2689 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2690 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2691 \1''', webpage)]
2692
2693 # lazyYT YouTube embed
2694 entries.extend(list(map(
2695 unescapeHTML,
2696 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2697
2698 # Wordpress "YouTube Video Importer" plugin
2699 matches = re.findall(r'''(?x)<div[^>]+
2700 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2701 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2702 entries.extend(m[-1] for m in matches)
2703
2704 return entries
2705
2706 @staticmethod
2707 def _extract_url(webpage):
2708 urls = YoutubeIE._extract_urls(webpage)
2709 return urls[0] if urls else None
2710
2711 @classmethod
2712 def extract_id(cls, url):
2713 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2714 if mobj is None:
2715 raise ExtractorError('Invalid URL: %s' % url)
2716 return mobj.group('id')
2717
2718 def _extract_chapters_from_json(self, data, duration):
2719 chapter_list = traverse_obj(
2720 data, (
2721 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2722 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2723 ), expected_type=list)
2724
2725 return self._extract_chapters(
2726 chapter_list,
2727 chapter_time=lambda chapter: float_or_none(
2728 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2729 chapter_title=lambda chapter: traverse_obj(
2730 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2731 duration=duration)
2732
2733 def _extract_chapters_from_engagement_panel(self, data, duration):
2734 content_list = traverse_obj(
2735 data,
2736 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2737 expected_type=list, default=[])
2738 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2739 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2740
2741 return next(filter(None, (
2742 self._extract_chapters(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2743 chapter_time, chapter_title, duration)
2744 for contents in content_list)), [])
2745
2746 def _extract_chapters_from_description(self, description, duration):
2747 return self._extract_chapters(
2748 re.findall(r'(?m)^((?:\d+:)?\d{1,2}:\d{2})\b\W*\s(.+?)\s*$', description or ''),
2749 chapter_time=lambda x: parse_duration(x[0]), chapter_title=lambda x: x[1],
2750 duration=duration, strict=False)
2751
2752 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration, strict=True):
2753 if not duration:
2754 return
2755 chapter_list = [{
2756 'start_time': chapter_time(chapter),
2757 'title': chapter_title(chapter),
2758 } for chapter in chapter_list or []]
2759 if not strict:
2760 chapter_list.sort(key=lambda c: c['start_time'] or 0)
2761
2762 chapters = [{'start_time': 0, 'title': '<Untitled>'}]
2763 for idx, chapter in enumerate(chapter_list):
2764 if chapter['start_time'] is None or not chapter['title']:
2765 self.report_warning(f'Incomplete chapter {idx}')
2766 elif chapters[-1]['start_time'] <= chapter['start_time'] <= duration:
2767 chapters[-1]['end_time'] = chapter['start_time']
2768 chapters.append(chapter)
2769 else:
2770 self.report_warning(f'Invalid start time for chapter "{chapter["title"]}"')
2771 chapters[-1]['end_time'] = duration
2772 return chapters if len(chapters) > 1 and chapters[1]['start_time'] else chapters[1:]
2773
2774 def _extract_comment(self, comment_renderer, parent=None):
2775 comment_id = comment_renderer.get('commentId')
2776 if not comment_id:
2777 return
2778
2779 text = self._get_text(comment_renderer, 'contentText')
2780
2781 # note: timestamp is an estimate calculated from the current time and time_text
2782 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2783 author = self._get_text(comment_renderer, 'authorText')
2784 author_id = try_get(comment_renderer,
2785 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2786
2787 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2788 lambda x: x['likeCount']), compat_str)) or 0
2789 author_thumbnail = try_get(comment_renderer,
2790 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2791
2792 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2793 is_favorited = 'creatorHeart' in (try_get(
2794 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2795 return {
2796 'id': comment_id,
2797 'text': text,
2798 'timestamp': timestamp,
2799 'time_text': time_text,
2800 'like_count': votes,
2801 'is_favorited': is_favorited,
2802 'author': author,
2803 'author_id': author_id,
2804 'author_thumbnail': author_thumbnail,
2805 'author_is_uploader': author_is_uploader,
2806 'parent': parent or 'root'
2807 }
2808
2809 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2810
2811 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2812
2813 def extract_header(contents):
2814 _continuation = None
2815 for content in contents:
2816 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2817 expected_comment_count = self._get_count(
2818 comments_header_renderer, 'countText', 'commentsCount')
2819
2820 if expected_comment_count:
2821 tracker['est_total'] = expected_comment_count
2822 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2823 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2824
2825 sort_menu_item = try_get(
2826 comments_header_renderer,
2827 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2828 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2829
2830 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2831 if not _continuation:
2832 continue
2833
2834 sort_text = str_or_none(sort_menu_item.get('title'))
2835 if not sort_text:
2836 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2837 self.to_screen('Sorting comments by %s' % sort_text.lower())
2838 break
2839 return _continuation
2840
2841 def extract_thread(contents):
2842 if not parent:
2843 tracker['current_page_thread'] = 0
2844 for content in contents:
2845 if not parent and tracker['total_parent_comments'] >= max_parents:
2846 yield
2847 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2848 comment_renderer = get_first(
2849 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2850 expected_type=dict, default={})
2851
2852 comment = self._extract_comment(comment_renderer, parent)
2853 if not comment:
2854 continue
2855
2856 tracker['running_total'] += 1
2857 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2858 yield comment
2859
2860 # Attempt to get the replies
2861 comment_replies_renderer = try_get(
2862 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2863
2864 if comment_replies_renderer:
2865 tracker['current_page_thread'] += 1
2866 comment_entries_iter = self._comment_entries(
2867 comment_replies_renderer, ytcfg, video_id,
2868 parent=comment.get('id'), tracker=tracker)
2869 yield from itertools.islice(comment_entries_iter, min(
2870 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2871
2872 # Keeps track of counts across recursive calls
2873 if not tracker:
2874 tracker = dict(
2875 running_total=0,
2876 est_total=0,
2877 current_page_thread=0,
2878 total_parent_comments=0,
2879 total_reply_comments=0)
2880
2881 # TODO: Deprecated
2882 # YouTube comments have a max depth of 2
2883 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2884 if max_depth:
2885 self._downloader.deprecation_warning(
2886 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2887 if max_depth == 1 and parent:
2888 return
2889
2890 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2891 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2892
2893 continuation = self._extract_continuation(root_continuation_data)
2894
2895 response = None
2896 is_forced_continuation = False
2897 is_first_continuation = parent is None
2898 if is_first_continuation and not continuation:
2899 # Sometimes you can get comments by generating the continuation yourself,
2900 # even if YouTube initially reports them being disabled - e.g. stories comments.
2901 # Note: if the comment section is actually disabled, YouTube may return a response with
2902 # required check_get_keys missing. So we will disable that check initially in this case.
2903 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2904 is_forced_continuation = True
2905
2906 for page_num in itertools.count(0):
2907 if not continuation:
2908 break
2909 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2910 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2911 if page_num == 0:
2912 if is_first_continuation:
2913 note_prefix = 'Downloading comment section API JSON'
2914 else:
2915 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2916 tracker['current_page_thread'], comment_prog_str)
2917 else:
2918 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2919 ' ' if parent else '', ' replies' if parent else '',
2920 page_num, comment_prog_str)
2921
2922 response = self._extract_response(
2923 item_id=None, query=continuation,
2924 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2925 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2926 is_forced_continuation = False
2927 continuation_contents = traverse_obj(
2928 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2929
2930 continuation = None
2931 for continuation_section in continuation_contents:
2932 continuation_items = traverse_obj(
2933 continuation_section,
2934 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2935 get_all=False, expected_type=list) or []
2936 if is_first_continuation:
2937 continuation = extract_header(continuation_items)
2938 is_first_continuation = False
2939 if continuation:
2940 break
2941 continue
2942
2943 for entry in extract_thread(continuation_items):
2944 if not entry:
2945 return
2946 yield entry
2947 continuation = self._extract_continuation({'contents': continuation_items})
2948 if continuation:
2949 break
2950
2951 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2952 if message and not parent and tracker['running_total'] == 0:
2953 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2954
2955 @staticmethod
2956 def _generate_comment_continuation(video_id):
2957 """
2958 Generates initial comment section continuation token from given video id
2959 """
2960 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2961 return base64.b64encode(token.encode()).decode()
2962
2963 def _get_comments(self, ytcfg, video_id, contents, webpage):
2964 """Entry for comment extraction"""
2965 def _real_comment_extract(contents):
2966 renderer = next((
2967 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2968 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2969 yield from self._comment_entries(renderer, ytcfg, video_id)
2970
2971 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2972 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2973
2974 @staticmethod
2975 def _get_checkok_params():
2976 return {'contentCheckOk': True, 'racyCheckOk': True}
2977
2978 @classmethod
2979 def _generate_player_context(cls, sts=None):
2980 context = {
2981 'html5Preference': 'HTML5_PREF_WANTS',
2982 }
2983 if sts is not None:
2984 context['signatureTimestamp'] = sts
2985 return {
2986 'playbackContext': {
2987 'contentPlaybackContext': context
2988 },
2989 **cls._get_checkok_params()
2990 }
2991
2992 @staticmethod
2993 def _is_agegated(player_response):
2994 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2995 return True
2996
2997 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2998 AGE_GATE_REASONS = (
2999 'confirm your age', 'age-restricted', 'inappropriate', # reason
3000 'age_verification_required', 'age_check_required', # status
3001 )
3002 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3003
3004 @staticmethod
3005 def _is_unplayable(player_response):
3006 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3007
3008 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
3009
3010 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3011 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3012 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3013 headers = self.generate_api_headers(
3014 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3015
3016 yt_query = {
3017 'videoId': video_id,
3018 'params': '8AEB' # enable stories
3019 }
3020 yt_query.update(self._generate_player_context(sts))
3021 return self._extract_response(
3022 item_id=video_id, ep='player', query=yt_query,
3023 ytcfg=player_ytcfg, headers=headers, fatal=True,
3024 default_client=client,
3025 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3026 ) or None
3027
3028 def _get_requested_clients(self, url, smuggled_data):
3029 requested_clients = []
3030 default = ['android', 'web']
3031 allowed_clients = sorted(
3032 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3033 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3034 for client in self._configuration_arg('player_client'):
3035 if client in allowed_clients:
3036 requested_clients.append(client)
3037 elif client == 'default':
3038 requested_clients.extend(default)
3039 elif client == 'all':
3040 requested_clients.extend(allowed_clients)
3041 else:
3042 self.report_warning(f'Skipping unsupported client {client}')
3043 if not requested_clients:
3044 requested_clients = default
3045
3046 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3047 requested_clients.extend(
3048 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3049
3050 return orderedSet(requested_clients)
3051
3052 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3053 initial_pr = None
3054 if webpage:
3055 initial_pr = self._extract_yt_initial_variable(
3056 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
3057 video_id, 'initial player response')
3058
3059 all_clients = set(clients)
3060 clients = clients[::-1]
3061 prs = []
3062
3063 def append_client(*client_names):
3064 """ Append the first client name that exists but not already used """
3065 for client_name in client_names:
3066 actual_client = _split_innertube_client(client_name)[0]
3067 if actual_client in INNERTUBE_CLIENTS:
3068 if actual_client not in all_clients:
3069 clients.append(client_name)
3070 all_clients.add(actual_client)
3071 return
3072
3073 # Android player_response does not have microFormats which are needed for
3074 # extraction of some data. So we return the initial_pr with formats
3075 # stripped out even if not requested by the user
3076 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3077 if initial_pr:
3078 pr = dict(initial_pr)
3079 pr['streamingData'] = None
3080 prs.append(pr)
3081
3082 last_error = None
3083 tried_iframe_fallback = False
3084 player_url = None
3085 while clients:
3086 client, base_client, variant = _split_innertube_client(clients.pop())
3087 player_ytcfg = master_ytcfg if client == 'web' else {}
3088 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3089 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3090
3091 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3092 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3093 if 'js' in self._configuration_arg('player_skip'):
3094 require_js_player = False
3095 player_url = None
3096
3097 if not player_url and not tried_iframe_fallback and require_js_player:
3098 player_url = self._download_player_url(video_id)
3099 tried_iframe_fallback = True
3100
3101 try:
3102 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3103 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3104 except ExtractorError as e:
3105 if last_error:
3106 self.report_warning(last_error)
3107 last_error = e
3108 continue
3109
3110 if pr:
3111 prs.append(pr)
3112
3113 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3114 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3115 append_client(f'{base_client}_creator')
3116 elif self._is_agegated(pr):
3117 if variant == 'tv_embedded':
3118 append_client(f'{base_client}_embedded')
3119 elif not variant:
3120 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3121
3122 if last_error:
3123 if not len(prs):
3124 raise last_error
3125 self.report_warning(last_error)
3126 return prs, player_url
3127
3128 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3129 itags, stream_ids = {}, []
3130 itag_qualities, res_qualities = {}, {}
3131 q = qualities([
3132 # Normally tiny is the smallest video-only formats. But
3133 # audio-only formats with unknown quality may get tagged as tiny
3134 'tiny',
3135 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3136 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3137 ])
3138 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3139
3140 for fmt in streaming_formats:
3141 if fmt.get('targetDurationSec'):
3142 continue
3143
3144 itag = str_or_none(fmt.get('itag'))
3145 audio_track = fmt.get('audioTrack') or {}
3146 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3147 if stream_id in stream_ids:
3148 continue
3149
3150 quality = fmt.get('quality')
3151 height = int_or_none(fmt.get('height'))
3152 if quality == 'tiny' or not quality:
3153 quality = fmt.get('audioQuality', '').lower() or quality
3154 # The 3gp format (17) in android client has a quality of "small",
3155 # but is actually worse than other formats
3156 if itag == '17':
3157 quality = 'tiny'
3158 if quality:
3159 if itag:
3160 itag_qualities[itag] = quality
3161 if height:
3162 res_qualities[height] = quality
3163 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3164 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3165 # number of fragment that would subsequently requested with (`&sq=N`)
3166 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3167 continue
3168
3169 fmt_url = fmt.get('url')
3170 if not fmt_url:
3171 sc = compat_parse_qs(fmt.get('signatureCipher'))
3172 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3173 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3174 if not all((sc, fmt_url, player_url, encrypted_sig)):
3175 continue
3176 try:
3177 fmt_url += '&%s=%s' % (
3178 traverse_obj(sc, ('sp', -1)) or 'signature',
3179 self._decrypt_signature(encrypted_sig, video_id, player_url)
3180 )
3181 except ExtractorError as e:
3182 self.report_warning('Signature extraction failed: Some formats may be missing', only_once=True)
3183 self.write_debug(e, only_once=True)
3184 continue
3185
3186 query = parse_qs(fmt_url)
3187 throttled = False
3188 if query.get('n'):
3189 try:
3190 fmt_url = update_url_query(fmt_url, {
3191 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3192 except ExtractorError as e:
3193 self.report_warning(
3194 'nsig extraction failed: You may experience throttling for some formats\n'
3195 f'n = {query["n"][0]} ; player = {player_url}', only_once=True)
3196 self.write_debug(e, only_once=True)
3197 throttled = True
3198
3199 if itag:
3200 itags[itag] = 'https'
3201 stream_ids.append(stream_id)
3202
3203 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3204 language_preference = (
3205 10 if audio_track.get('audioIsDefault') and 10
3206 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3207 else -1)
3208 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3209 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3210 # Make sure to avoid false positives with small duration differences.
3211 # Eg: __2ABJjxzNo, ySuUZEjARPY
3212 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3213 if is_damaged:
3214 self.report_warning(
3215 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3216 dct = {
3217 'asr': int_or_none(fmt.get('audioSampleRate')),
3218 'filesize': int_or_none(fmt.get('contentLength')),
3219 'format_id': itag,
3220 'format_note': join_nonempty(
3221 '%s%s' % (audio_track.get('displayName') or '',
3222 ' (default)' if language_preference > 0 else ''),
3223 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3224 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3225 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3226 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3227 'fps': int_or_none(fmt.get('fps')) or None,
3228 'height': height,
3229 'quality': q(quality),
3230 'has_drm': bool(fmt.get('drmFamilies')),
3231 'tbr': tbr,
3232 'url': fmt_url,
3233 'width': int_or_none(fmt.get('width')),
3234 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3235 'desc' if language_preference < -1 else ''),
3236 'language_preference': language_preference,
3237 # Strictly de-prioritize damaged and 3gp formats
3238 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3239 }
3240 mime_mobj = re.match(
3241 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3242 if mime_mobj:
3243 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3244 dct.update(parse_codecs(mime_mobj.group(2)))
3245 no_audio = dct.get('acodec') == 'none'
3246 no_video = dct.get('vcodec') == 'none'
3247 if no_audio:
3248 dct['vbr'] = tbr
3249 if no_video:
3250 dct['abr'] = tbr
3251 if no_audio or no_video:
3252 dct['downloader_options'] = {
3253 # Youtube throttles chunks >~10M
3254 'http_chunk_size': 10485760,
3255 }
3256 if dct.get('ext'):
3257 dct['container'] = dct['ext'] + '_dash'
3258 yield dct
3259
3260 live_from_start = is_live and self.get_param('live_from_start')
3261 skip_manifests = self._configuration_arg('skip')
3262 if not self.get_param('youtube_include_hls_manifest', True):
3263 skip_manifests.append('hls')
3264 if not self.get_param('youtube_include_dash_manifest', True):
3265 skip_manifests.append('dash')
3266 get_dash = 'dash' not in skip_manifests and (
3267 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3268 get_hls = not live_from_start and 'hls' not in skip_manifests
3269
3270 def process_manifest_format(f, proto, itag):
3271 if itag in itags:
3272 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3273 return False
3274 itag = f'{itag}-{proto}'
3275 if itag:
3276 f['format_id'] = itag
3277 itags[itag] = proto
3278
3279 f['quality'] = next((
3280 q(qdict[val])
3281 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3282 if val in qdict), -1)
3283 return True
3284
3285 for sd in streaming_data:
3286 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3287 if hls_manifest_url:
3288 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3289 if process_manifest_format(f, 'hls', self._search_regex(
3290 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3291 yield f
3292
3293 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3294 if dash_manifest_url:
3295 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3296 if process_manifest_format(f, 'dash', f['format_id']):
3297 f['filesize'] = int_or_none(self._search_regex(
3298 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3299 if live_from_start:
3300 f['is_from_start'] = True
3301
3302 yield f
3303
3304 def _extract_storyboard(self, player_responses, duration):
3305 spec = get_first(
3306 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3307 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3308 if not base_url:
3309 return
3310 L = len(spec) - 1
3311 for i, args in enumerate(spec):
3312 args = args.split('#')
3313 counts = list(map(int_or_none, args[:5]))
3314 if len(args) != 8 or not all(counts):
3315 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3316 continue
3317 width, height, frame_count, cols, rows = counts
3318 N, sigh = args[6:]
3319
3320 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3321 fragment_count = frame_count / (cols * rows)
3322 fragment_duration = duration / fragment_count
3323 yield {
3324 'format_id': f'sb{i}',
3325 'format_note': 'storyboard',
3326 'ext': 'mhtml',
3327 'protocol': 'mhtml',
3328 'acodec': 'none',
3329 'vcodec': 'none',
3330 'url': url,
3331 'width': width,
3332 'height': height,
3333 'fragments': [{
3334 'url': url.replace('$M', str(j)),
3335 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3336 } for j in range(math.ceil(fragment_count))],
3337 }
3338
3339 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3340 webpage = None
3341 if 'webpage' not in self._configuration_arg('player_skip'):
3342 webpage = self._download_webpage(
3343 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3344
3345 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3346
3347 player_responses, player_url = self._extract_player_responses(
3348 self._get_requested_clients(url, smuggled_data),
3349 video_id, webpage, master_ytcfg)
3350
3351 return webpage, master_ytcfg, player_responses, player_url
3352
3353 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3354 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3355 is_live = get_first(video_details, 'isLive')
3356 if is_live is None:
3357 is_live = get_first(live_broadcast_details, 'isLiveNow')
3358
3359 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3360 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3361
3362 return live_broadcast_details, is_live, streaming_data, formats
3363
3364 def _real_extract(self, url):
3365 url, smuggled_data = unsmuggle_url(url, {})
3366 video_id = self._match_id(url)
3367
3368 base_url = self.http_scheme() + '//www.youtube.com/'
3369 webpage_url = base_url + 'watch?v=' + video_id
3370
3371 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3372
3373 playability_statuses = traverse_obj(
3374 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3375
3376 trailer_video_id = get_first(
3377 playability_statuses,
3378 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3379 expected_type=str)
3380 if trailer_video_id:
3381 return self.url_result(
3382 trailer_video_id, self.ie_key(), trailer_video_id)
3383
3384 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3385 if webpage else (lambda x: None))
3386
3387 video_details = traverse_obj(
3388 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3389 microformats = traverse_obj(
3390 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3391 expected_type=dict, default=[])
3392 video_title = (
3393 get_first(video_details, 'title')
3394 or self._get_text(microformats, (..., 'title'))
3395 or search_meta(['og:title', 'twitter:title', 'title']))
3396 video_description = get_first(video_details, 'shortDescription')
3397
3398 multifeed_metadata_list = get_first(
3399 player_responses,
3400 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3401 expected_type=str)
3402 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3403 if self.get_param('noplaylist'):
3404 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3405 else:
3406 entries = []
3407 feed_ids = []
3408 for feed in multifeed_metadata_list.split(','):
3409 # Unquote should take place before split on comma (,) since textual
3410 # fields may contain comma as well (see
3411 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3412 feed_data = compat_parse_qs(
3413 compat_urllib_parse_unquote_plus(feed))
3414
3415 def feed_entry(name):
3416 return try_get(
3417 feed_data, lambda x: x[name][0], compat_str)
3418
3419 feed_id = feed_entry('id')
3420 if not feed_id:
3421 continue
3422 feed_title = feed_entry('title')
3423 title = video_title
3424 if feed_title:
3425 title += ' (%s)' % feed_title
3426 entries.append({
3427 '_type': 'url_transparent',
3428 'ie_key': 'Youtube',
3429 'url': smuggle_url(
3430 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3431 {'force_singlefeed': True}),
3432 'title': title,
3433 })
3434 feed_ids.append(feed_id)
3435 self.to_screen(
3436 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3437 % (', '.join(feed_ids), video_id))
3438 return self.playlist_result(
3439 entries, video_id, video_title, video_description)
3440
3441 duration = int_or_none(
3442 get_first(video_details, 'lengthSeconds')
3443 or get_first(microformats, 'lengthSeconds')
3444 or parse_duration(search_meta('duration'))) or None
3445
3446 if get_first(video_details, 'isPostLiveDvr'):
3447 self.write_debug('Video is in Post-Live Manifestless mode')
3448 if duration or 0 > 4 * 3600:
3449 self.report_warning(
3450 'The livestream has not finished processing. Only 4 hours of the video can be currently downloaded. '
3451 'This is a known issue and patches are welcome')
3452
3453 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3454 video_id, microformats, video_details, player_responses, player_url, duration)
3455
3456 if not formats:
3457 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3458 self.report_drm(video_id)
3459 pemr = get_first(
3460 playability_statuses,
3461 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3462 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3463 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3464 if subreason:
3465 if subreason == 'The uploader has not made this video available in your country.':
3466 countries = get_first(microformats, 'availableCountries')
3467 if not countries:
3468 regions_allowed = search_meta('regionsAllowed')
3469 countries = regions_allowed.split(',') if regions_allowed else None
3470 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3471 reason += f'. {subreason}'
3472 if reason:
3473 self.raise_no_formats(reason, expected=True)
3474
3475 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3476 if not keywords and webpage:
3477 keywords = [
3478 unescapeHTML(m.group('content'))
3479 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3480 for keyword in keywords:
3481 if keyword.startswith('yt:stretch='):
3482 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3483 if mobj:
3484 # NB: float is intentional for forcing float division
3485 w, h = (float(v) for v in mobj.groups())
3486 if w > 0 and h > 0:
3487 ratio = w / h
3488 for f in formats:
3489 if f.get('vcodec') != 'none':
3490 f['stretched_ratio'] = ratio
3491 break
3492 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3493 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3494 if thumbnail_url:
3495 thumbnails.append({
3496 'url': thumbnail_url,
3497 })
3498 original_thumbnails = thumbnails.copy()
3499
3500 # The best resolution thumbnails sometimes does not appear in the webpage
3501 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3502 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3503 thumbnail_names = [
3504 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3505 # in resolution, these are not the custom thumbnail. So de-prioritize them
3506 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3507 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3508 ]
3509 n_thumbnail_names = len(thumbnail_names)
3510 thumbnails.extend({
3511 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3512 video_id=video_id, name=name, ext=ext,
3513 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3514 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3515 for thumb in thumbnails:
3516 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3517 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3518 self._remove_duplicate_formats(thumbnails)
3519 self._downloader._sort_thumbnails(original_thumbnails)
3520
3521 category = get_first(microformats, 'category') or search_meta('genre')
3522 channel_id = str_or_none(
3523 get_first(video_details, 'channelId')
3524 or get_first(microformats, 'externalChannelId')
3525 or search_meta('channelId'))
3526 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3527
3528 live_content = get_first(video_details, 'isLiveContent')
3529 is_upcoming = get_first(video_details, 'isUpcoming')
3530 if is_live is None:
3531 if is_upcoming or live_content is False:
3532 is_live = False
3533 if is_upcoming is None and (live_content or is_live):
3534 is_upcoming = False
3535 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3536 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3537 if not duration and live_end_time and live_start_time:
3538 duration = live_end_time - live_start_time
3539
3540 if is_live and self.get_param('live_from_start'):
3541 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3542
3543 formats.extend(self._extract_storyboard(player_responses, duration))
3544
3545 # Source is given priority since formats that throttle are given lower source_preference
3546 # When throttling issue is fully fixed, remove this
3547 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3548
3549 info = {
3550 'id': video_id,
3551 'title': video_title,
3552 'formats': formats,
3553 'thumbnails': thumbnails,
3554 # The best thumbnail that we are sure exists. Prevents unnecessary
3555 # URL checking if user don't care about getting the best possible thumbnail
3556 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3557 'description': video_description,
3558 'uploader': get_first(video_details, 'author'),
3559 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3560 'uploader_url': owner_profile_url,
3561 'channel_id': channel_id,
3562 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3563 'duration': duration,
3564 'view_count': int_or_none(
3565 get_first((video_details, microformats), (..., 'viewCount'))
3566 or search_meta('interactionCount')),
3567 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3568 'age_limit': 18 if (
3569 get_first(microformats, 'isFamilySafe') is False
3570 or search_meta('isFamilyFriendly') == 'false'
3571 or search_meta('og:restrictions:age') == '18+') else 0,
3572 'webpage_url': webpage_url,
3573 'categories': [category] if category else None,
3574 'tags': keywords,
3575 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3576 'is_live': is_live,
3577 'was_live': (False if is_live or is_upcoming or live_content is False
3578 else None if is_live is None or is_upcoming is None
3579 else live_content),
3580 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3581 'release_timestamp': live_start_time,
3582 }
3583
3584 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3585 if pctr:
3586 def get_lang_code(track):
3587 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3588 or track.get('languageCode'))
3589
3590 # Converted into dicts to remove duplicates
3591 captions = {
3592 get_lang_code(sub): sub
3593 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3594 translation_languages = {
3595 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3596 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3597
3598 def process_language(container, base_url, lang_code, sub_name, query):
3599 lang_subs = container.setdefault(lang_code, [])
3600 for fmt in self._SUBTITLE_FORMATS:
3601 query.update({
3602 'fmt': fmt,
3603 })
3604 lang_subs.append({
3605 'ext': fmt,
3606 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3607 'name': sub_name,
3608 })
3609
3610 subtitles, automatic_captions = {}, {}
3611 for lang_code, caption_track in captions.items():
3612 base_url = caption_track.get('baseUrl')
3613 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3614 if not base_url:
3615 continue
3616 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3617 if caption_track.get('kind') != 'asr':
3618 if not lang_code:
3619 continue
3620 process_language(
3621 subtitles, base_url, lang_code, lang_name, {})
3622 if not caption_track.get('isTranslatable'):
3623 continue
3624 for trans_code, trans_name in translation_languages.items():
3625 if not trans_code:
3626 continue
3627 orig_trans_code = trans_code
3628 if caption_track.get('kind') != 'asr':
3629 if 'translated_subs' in self._configuration_arg('skip'):
3630 continue
3631 trans_code += f'-{lang_code}'
3632 trans_name += format_field(lang_name, template=' from %s')
3633 # Add an "-orig" label to the original language so that it can be distinguished.
3634 # The subs are returned without "-orig" as well for compatibility
3635 if lang_code == f'a-{orig_trans_code}':
3636 process_language(
3637 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3638 # Setting tlang=lang returns damaged subtitles.
3639 process_language(automatic_captions, base_url, trans_code, trans_name,
3640 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3641 info['automatic_captions'] = automatic_captions
3642 info['subtitles'] = subtitles
3643
3644 parsed_url = compat_urllib_parse_urlparse(url)
3645 for component in [parsed_url.fragment, parsed_url.query]:
3646 query = compat_parse_qs(component)
3647 for k, v in query.items():
3648 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3649 d_k += '_time'
3650 if d_k not in info and k in s_ks:
3651 info[d_k] = parse_duration(query[k][0])
3652
3653 # Youtube Music Auto-generated description
3654 if video_description:
3655 mobj = re.search(
3656 r'''(?xs)
3657 (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
3658 (?P<album>[^\n]+)
3659 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
3660 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
3661 (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
3662 .+\nAuto-generated\ by\ YouTube\.\s*$
3663 ''', video_description)
3664 if mobj:
3665 release_year = mobj.group('release_year')
3666 release_date = mobj.group('release_date')
3667 if release_date:
3668 release_date = release_date.replace('-', '')
3669 if not release_year:
3670 release_year = release_date[:4]
3671 info.update({
3672 'album': mobj.group('album'.strip()),
3673 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3674 'track': mobj.group('track').strip(),
3675 'release_date': release_date,
3676 'release_year': int_or_none(release_year),
3677 })
3678
3679 initial_data = None
3680 if webpage:
3681 initial_data = self._extract_yt_initial_variable(
3682 webpage, self._YT_INITIAL_DATA_RE, video_id,
3683 'yt initial data')
3684 if not initial_data:
3685 query = {'videoId': video_id}
3686 query.update(self._get_checkok_params())
3687 initial_data = self._extract_response(
3688 item_id=video_id, ep='next', fatal=False,
3689 ytcfg=master_ytcfg, query=query,
3690 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3691 note='Downloading initial data API JSON')
3692
3693 try: # This will error if there is no livechat
3694 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3695 except (KeyError, IndexError, TypeError):
3696 pass
3697 else:
3698 info.setdefault('subtitles', {})['live_chat'] = [{
3699 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3700 'video_id': video_id,
3701 'ext': 'json',
3702 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3703 }]
3704
3705 if initial_data:
3706 info['chapters'] = (
3707 self._extract_chapters_from_json(initial_data, duration)
3708 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3709 or self._extract_chapters_from_description(video_description, duration)
3710 or None)
3711
3712 contents = traverse_obj(
3713 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3714 expected_type=list, default=[])
3715
3716 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3717 if vpir:
3718 stl = vpir.get('superTitleLink')
3719 if stl:
3720 stl = self._get_text(stl)
3721 if try_get(
3722 vpir,
3723 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3724 info['location'] = stl
3725 else:
3726 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3727 if mobj:
3728 info.update({
3729 'series': mobj.group(1),
3730 'season_number': int(mobj.group(2)),
3731 'episode_number': int(mobj.group(3)),
3732 })
3733 for tlb in (try_get(
3734 vpir,
3735 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3736 list) or []):
3737 tbr = tlb.get('toggleButtonRenderer') or {}
3738 for getter, regex in [(
3739 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3740 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3741 lambda x: x['accessibility'],
3742 lambda x: x['accessibilityData']['accessibilityData'],
3743 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3744 label = (try_get(tbr, getter, dict) or {}).get('label')
3745 if label:
3746 mobj = re.match(regex, label)
3747 if mobj:
3748 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3749 break
3750 sbr_tooltip = try_get(
3751 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3752 if sbr_tooltip:
3753 like_count, dislike_count = sbr_tooltip.split(' / ')
3754 info.update({
3755 'like_count': str_to_int(like_count),
3756 'dislike_count': str_to_int(dislike_count),
3757 })
3758 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3759 if vsir:
3760 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3761 info.update({
3762 'channel': self._get_text(vor, 'title'),
3763 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3764
3765 rows = try_get(
3766 vsir,
3767 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3768 list) or []
3769 multiple_songs = False
3770 for row in rows:
3771 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3772 multiple_songs = True
3773 break
3774 for row in rows:
3775 mrr = row.get('metadataRowRenderer') or {}
3776 mrr_title = mrr.get('title')
3777 if not mrr_title:
3778 continue
3779 mrr_title = self._get_text(mrr, 'title')
3780 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3781 if mrr_title == 'License':
3782 info['license'] = mrr_contents_text
3783 elif not multiple_songs:
3784 if mrr_title == 'Album':
3785 info['album'] = mrr_contents_text
3786 elif mrr_title == 'Artist':
3787 info['artist'] = mrr_contents_text
3788 elif mrr_title == 'Song':
3789 info['track'] = mrr_contents_text
3790
3791 fallbacks = {
3792 'channel': 'uploader',
3793 'channel_id': 'uploader_id',
3794 'channel_url': 'uploader_url',
3795 }
3796
3797 # The upload date for scheduled, live and past live streams / premieres in microformats
3798 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3799 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3800 upload_date = (
3801 unified_strdate(get_first(microformats, 'uploadDate'))
3802 or unified_strdate(search_meta('uploadDate')))
3803 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3804 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3805 info['upload_date'] = upload_date
3806
3807 for to, frm in fallbacks.items():
3808 if not info.get(to):
3809 info[to] = info.get(frm)
3810
3811 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3812 v = info.get(s_k)
3813 if v:
3814 info[d_k] = v
3815
3816 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3817 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3818 is_membersonly = None
3819 is_premium = None
3820 if initial_data and is_private is not None:
3821 is_membersonly = False
3822 is_premium = False
3823 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3824 badge_labels = set()
3825 for content in contents:
3826 if not isinstance(content, dict):
3827 continue
3828 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3829 for badge_label in badge_labels:
3830 if badge_label.lower() == 'members only':
3831 is_membersonly = True
3832 elif badge_label.lower() == 'premium':
3833 is_premium = True
3834 elif badge_label.lower() == 'unlisted':
3835 is_unlisted = True
3836
3837 info['availability'] = self._availability(
3838 is_private=is_private,
3839 needs_premium=is_premium,
3840 needs_subscription=is_membersonly,
3841 needs_auth=info['age_limit'] >= 18,
3842 is_unlisted=None if is_private is None else is_unlisted)
3843
3844 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3845
3846 self.mark_watched(video_id, player_responses)
3847
3848 return info
3849
3850
3851class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3852
3853 @staticmethod
3854 def passthrough_smuggled_data(func):
3855 def _smuggle(entries, smuggled_data):
3856 for entry in entries:
3857 # TODO: Convert URL to music.youtube instead.
3858 # Do we need to passthrough any other smuggled_data?
3859 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3860 yield entry
3861
3862 @functools.wraps(func)
3863 def wrapper(self, url):
3864 url, smuggled_data = unsmuggle_url(url, {})
3865 if self.is_music_url(url):
3866 smuggled_data['is_music_url'] = True
3867 info_dict = func(self, url, smuggled_data)
3868 if smuggled_data and info_dict.get('entries'):
3869 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3870 return info_dict
3871 return wrapper
3872
3873 def _extract_channel_id(self, webpage):
3874 channel_id = self._html_search_meta(
3875 'channelId', webpage, 'channel id', default=None)
3876 if channel_id:
3877 return channel_id
3878 channel_url = self._html_search_meta(
3879 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3880 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3881 'twitter:app:url:googleplay'), webpage, 'channel url')
3882 return self._search_regex(
3883 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3884 channel_url, 'channel id')
3885
3886 @staticmethod
3887 def _extract_basic_item_renderer(item):
3888 # Modified from _extract_grid_item_renderer
3889 known_basic_renderers = (
3890 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3891 )
3892 for key, renderer in item.items():
3893 if not isinstance(renderer, dict):
3894 continue
3895 elif key in known_basic_renderers:
3896 return renderer
3897 elif key.startswith('grid') and key.endswith('Renderer'):
3898 return renderer
3899
3900 def _grid_entries(self, grid_renderer):
3901 for item in grid_renderer['items']:
3902 if not isinstance(item, dict):
3903 continue
3904 renderer = self._extract_basic_item_renderer(item)
3905 if not isinstance(renderer, dict):
3906 continue
3907 title = self._get_text(renderer, 'title')
3908
3909 # playlist
3910 playlist_id = renderer.get('playlistId')
3911 if playlist_id:
3912 yield self.url_result(
3913 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3914 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3915 video_title=title)
3916 continue
3917 # video
3918 video_id = renderer.get('videoId')
3919 if video_id:
3920 yield self._extract_video(renderer)
3921 continue
3922 # channel
3923 channel_id = renderer.get('channelId')
3924 if channel_id:
3925 yield self.url_result(
3926 'https://www.youtube.com/channel/%s' % channel_id,
3927 ie=YoutubeTabIE.ie_key(), video_title=title)
3928 continue
3929 # generic endpoint URL support
3930 ep_url = urljoin('https://www.youtube.com/', try_get(
3931 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3932 compat_str))
3933 if ep_url:
3934 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3935 if ie.suitable(ep_url):
3936 yield self.url_result(
3937 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3938 break
3939
3940 def _music_reponsive_list_entry(self, renderer):
3941 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3942 if video_id:
3943 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3944 ie=YoutubeIE.ie_key(), video_id=video_id)
3945 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3946 if playlist_id:
3947 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3948 if video_id:
3949 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3950 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3951 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3952 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3953 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3954 if browse_id:
3955 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3956 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3957
3958 def _shelf_entries_from_content(self, shelf_renderer):
3959 content = shelf_renderer.get('content')
3960 if not isinstance(content, dict):
3961 return
3962 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3963 if renderer:
3964 # TODO: add support for nested playlists so each shelf is processed
3965 # as separate playlist
3966 # TODO: this includes only first N items
3967 yield from self._grid_entries(renderer)
3968 renderer = content.get('horizontalListRenderer')
3969 if renderer:
3970 # TODO
3971 pass
3972
3973 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3974 ep = try_get(
3975 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3976 compat_str)
3977 shelf_url = urljoin('https://www.youtube.com', ep)
3978 if shelf_url:
3979 # Skipping links to another channels, note that checking for
3980 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3981 # will not work
3982 if skip_channels and '/channels?' in shelf_url:
3983 return
3984 title = self._get_text(shelf_renderer, 'title')
3985 yield self.url_result(shelf_url, video_title=title)
3986 # Shelf may not contain shelf URL, fallback to extraction from content
3987 yield from self._shelf_entries_from_content(shelf_renderer)
3988
3989 def _playlist_entries(self, video_list_renderer):
3990 for content in video_list_renderer['contents']:
3991 if not isinstance(content, dict):
3992 continue
3993 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3994 if not isinstance(renderer, dict):
3995 continue
3996 video_id = renderer.get('videoId')
3997 if not video_id:
3998 continue
3999 yield self._extract_video(renderer)
4000
4001 def _rich_entries(self, rich_grid_renderer):
4002 renderer = try_get(
4003 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
4004 video_id = renderer.get('videoId')
4005 if not video_id:
4006 return
4007 yield self._extract_video(renderer)
4008
4009 def _video_entry(self, video_renderer):
4010 video_id = video_renderer.get('videoId')
4011 if video_id:
4012 return self._extract_video(video_renderer)
4013
4014 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4015 url = urljoin('https://youtube.com', traverse_obj(
4016 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4017 if url:
4018 return self.url_result(
4019 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4020
4021 def _post_thread_entries(self, post_thread_renderer):
4022 post_renderer = try_get(
4023 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4024 if not post_renderer:
4025 return
4026 # video attachment
4027 video_renderer = try_get(
4028 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4029 video_id = video_renderer.get('videoId')
4030 if video_id:
4031 entry = self._extract_video(video_renderer)
4032 if entry:
4033 yield entry
4034 # playlist attachment
4035 playlist_id = try_get(
4036 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
4037 if playlist_id:
4038 yield self.url_result(
4039 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4040 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4041 # inline video links
4042 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4043 for run in runs:
4044 if not isinstance(run, dict):
4045 continue
4046 ep_url = try_get(
4047 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
4048 if not ep_url:
4049 continue
4050 if not YoutubeIE.suitable(ep_url):
4051 continue
4052 ep_video_id = YoutubeIE._match_id(ep_url)
4053 if video_id == ep_video_id:
4054 continue
4055 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4056
4057 def _post_thread_continuation_entries(self, post_thread_continuation):
4058 contents = post_thread_continuation.get('contents')
4059 if not isinstance(contents, list):
4060 return
4061 for content in contents:
4062 renderer = content.get('backstagePostThreadRenderer')
4063 if isinstance(renderer, dict):
4064 yield from self._post_thread_entries(renderer)
4065 continue
4066 renderer = content.get('videoRenderer')
4067 if isinstance(renderer, dict):
4068 yield self._video_entry(renderer)
4069
4070 r''' # unused
4071 def _rich_grid_entries(self, contents):
4072 for content in contents:
4073 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4074 if video_renderer:
4075 entry = self._video_entry(video_renderer)
4076 if entry:
4077 yield entry
4078 '''
4079
4080 def _extract_entries(self, parent_renderer, continuation_list):
4081 # continuation_list is modified in-place with continuation_list = [continuation_token]
4082 continuation_list[:] = [None]
4083 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4084 for content in contents:
4085 if not isinstance(content, dict):
4086 continue
4087 is_renderer = traverse_obj(
4088 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4089 expected_type=dict)
4090 if not is_renderer:
4091 renderer = content.get('richItemRenderer')
4092 if renderer:
4093 for entry in self._rich_entries(renderer):
4094 yield entry
4095 continuation_list[0] = self._extract_continuation(parent_renderer)
4096 continue
4097 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4098 for isr_content in isr_contents:
4099 if not isinstance(isr_content, dict):
4100 continue
4101
4102 known_renderers = {
4103 'playlistVideoListRenderer': self._playlist_entries,
4104 'gridRenderer': self._grid_entries,
4105 'reelShelfRenderer': self._grid_entries,
4106 'shelfRenderer': self._shelf_entries,
4107 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4108 'backstagePostThreadRenderer': self._post_thread_entries,
4109 'videoRenderer': lambda x: [self._video_entry(x)],
4110 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4111 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4112 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4113 }
4114 for key, renderer in isr_content.items():
4115 if key not in known_renderers:
4116 continue
4117 for entry in known_renderers[key](renderer):
4118 if entry:
4119 yield entry
4120 continuation_list[0] = self._extract_continuation(renderer)
4121 break
4122
4123 if not continuation_list[0]:
4124 continuation_list[0] = self._extract_continuation(is_renderer)
4125
4126 if not continuation_list[0]:
4127 continuation_list[0] = self._extract_continuation(parent_renderer)
4128
4129 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4130 continuation_list = [None]
4131 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4132 tab_content = try_get(tab, lambda x: x['content'], dict)
4133 if not tab_content:
4134 return
4135 parent_renderer = (
4136 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4137 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4138 yield from extract_entries(parent_renderer)
4139 continuation = continuation_list[0]
4140
4141 for page_num in itertools.count(1):
4142 if not continuation:
4143 break
4144 headers = self.generate_api_headers(
4145 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4146 response = self._extract_response(
4147 item_id=f'{item_id} page {page_num}',
4148 query=continuation, headers=headers, ytcfg=ytcfg,
4149 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4150
4151 if not response:
4152 break
4153 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4154 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4155 visitor_data = self._extract_visitor_data(response) or visitor_data
4156
4157 known_continuation_renderers = {
4158 'playlistVideoListContinuation': self._playlist_entries,
4159 'gridContinuation': self._grid_entries,
4160 'itemSectionContinuation': self._post_thread_continuation_entries,
4161 'sectionListContinuation': extract_entries, # for feeds
4162 }
4163 continuation_contents = try_get(
4164 response, lambda x: x['continuationContents'], dict) or {}
4165 continuation_renderer = None
4166 for key, value in continuation_contents.items():
4167 if key not in known_continuation_renderers:
4168 continue
4169 continuation_renderer = value
4170 continuation_list = [None]
4171 yield from known_continuation_renderers[key](continuation_renderer)
4172 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4173 break
4174 if continuation_renderer:
4175 continue
4176
4177 known_renderers = {
4178 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4179 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4180 'gridVideoRenderer': (self._grid_entries, 'items'),
4181 'gridChannelRenderer': (self._grid_entries, 'items'),
4182 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4183 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4184 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4185 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4186 }
4187 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4188 continuation_items = try_get(
4189 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4190 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4191 video_items_renderer = None
4192 for key, value in continuation_item.items():
4193 if key not in known_renderers:
4194 continue
4195 video_items_renderer = {known_renderers[key][1]: continuation_items}
4196 continuation_list = [None]
4197 yield from known_renderers[key][0](video_items_renderer)
4198 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4199 break
4200 if video_items_renderer:
4201 continue
4202 break
4203
4204 @staticmethod
4205 def _extract_selected_tab(tabs, fatal=True):
4206 for tab in tabs:
4207 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4208 if renderer.get('selected') is True:
4209 return renderer
4210 else:
4211 if fatal:
4212 raise ExtractorError('Unable to find selected tab')
4213
4214 def _extract_uploader(self, data):
4215 uploader = {}
4216 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4217 owner = try_get(
4218 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4219 if owner:
4220 owner_text = owner.get('text')
4221 uploader['uploader'] = self._search_regex(
4222 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4223 uploader['uploader_id'] = try_get(
4224 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4225 uploader['uploader_url'] = urljoin(
4226 'https://www.youtube.com/',
4227 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4228 return {k: v for k, v in uploader.items() if v is not None}
4229
4230 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4231 playlist_id = title = description = channel_url = channel_name = channel_id = None
4232 tags = []
4233
4234 selected_tab = self._extract_selected_tab(tabs)
4235 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4236 renderer = try_get(
4237 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4238 if renderer:
4239 channel_name = renderer.get('title')
4240 channel_url = renderer.get('channelUrl')
4241 channel_id = renderer.get('externalId')
4242 else:
4243 renderer = try_get(
4244 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4245
4246 if renderer:
4247 title = renderer.get('title')
4248 description = renderer.get('description', '')
4249 playlist_id = channel_id
4250 tags = renderer.get('keywords', '').split()
4251
4252 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4253 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4254 def _get_uncropped(url):
4255 return url_or_none((url or '').split('=')[0] + '=s0')
4256
4257 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4258 if avatar_thumbnails:
4259 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4260 if uncropped_avatar:
4261 avatar_thumbnails.append({
4262 'url': uncropped_avatar,
4263 'id': 'avatar_uncropped',
4264 'preference': 1
4265 })
4266
4267 channel_banners = self._extract_thumbnails(
4268 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4269 for banner in channel_banners:
4270 banner['preference'] = -10
4271
4272 if channel_banners:
4273 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4274 if uncropped_banner:
4275 channel_banners.append({
4276 'url': uncropped_banner,
4277 'id': 'banner_uncropped',
4278 'preference': -5
4279 })
4280
4281 primary_thumbnails = self._extract_thumbnails(
4282 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4283
4284 if playlist_id is None:
4285 playlist_id = item_id
4286
4287 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4288 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4289 if title is None:
4290 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4291 title += format_field(selected_tab, 'title', ' - %s')
4292 title += format_field(selected_tab, 'expandedText', ' - %s')
4293
4294 metadata = {
4295 'playlist_id': playlist_id,
4296 'playlist_title': title,
4297 'playlist_description': description,
4298 'uploader': channel_name,
4299 'uploader_id': channel_id,
4300 'uploader_url': channel_url,
4301 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4302 'tags': tags,
4303 'view_count': self._get_count(playlist_stats, 1),
4304 'availability': self._extract_availability(data),
4305 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4306 'playlist_count': self._get_count(playlist_stats, 0),
4307 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4308 }
4309 if not channel_id:
4310 metadata.update(self._extract_uploader(data))
4311 metadata.update({
4312 'channel': metadata['uploader'],
4313 'channel_id': metadata['uploader_id'],
4314 'channel_url': metadata['uploader_url']})
4315 return self.playlist_result(
4316 self._entries(
4317 selected_tab, playlist_id, ytcfg,
4318 self._extract_account_syncid(ytcfg, data),
4319 self._extract_visitor_data(data, ytcfg)),
4320 **metadata)
4321
4322 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4323 first_id = last_id = response = None
4324 for page_num in itertools.count(1):
4325 videos = list(self._playlist_entries(playlist))
4326 if not videos:
4327 return
4328 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4329 if start >= len(videos):
4330 return
4331 yield from videos[start:]
4332 first_id = first_id or videos[0]['id']
4333 last_id = videos[-1]['id']
4334 watch_endpoint = try_get(
4335 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4336 headers = self.generate_api_headers(
4337 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4338 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4339 query = {
4340 'playlistId': playlist_id,
4341 'videoId': watch_endpoint.get('videoId') or last_id,
4342 'index': watch_endpoint.get('index') or len(videos),
4343 'params': watch_endpoint.get('params') or 'OAE%3D'
4344 }
4345 response = self._extract_response(
4346 item_id='%s page %d' % (playlist_id, page_num),
4347 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4348 check_get_keys='contents'
4349 )
4350 playlist = try_get(
4351 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4352
4353 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4354 title = playlist.get('title') or try_get(
4355 data, lambda x: x['titleText']['simpleText'], compat_str)
4356 playlist_id = playlist.get('playlistId') or item_id
4357
4358 # Delegating everything except mix playlists to regular tab-based playlist URL
4359 playlist_url = urljoin(url, try_get(
4360 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4361 compat_str))
4362
4363 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4364 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4365 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4366
4367 if playlist_url and playlist_url != url and not is_known_unviewable:
4368 return self.url_result(
4369 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4370 video_title=title)
4371
4372 return self.playlist_result(
4373 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4374 playlist_id=playlist_id, playlist_title=title)
4375
4376 def _extract_availability(self, data):
4377 """
4378 Gets the availability of a given playlist/tab.
4379 Note: Unless YouTube tells us explicitly, we do not assume it is public
4380 @param data: response
4381 """
4382 is_private = is_unlisted = None
4383 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4384 badge_labels = self._extract_badges(renderer)
4385
4386 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4387 privacy_dropdown_entries = try_get(
4388 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4389 for renderer_dict in privacy_dropdown_entries:
4390 is_selected = try_get(
4391 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4392 if not is_selected:
4393 continue
4394 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4395 if label:
4396 badge_labels.add(label.lower())
4397 break
4398
4399 for badge_label in badge_labels:
4400 if badge_label == 'unlisted':
4401 is_unlisted = True
4402 elif badge_label == 'private':
4403 is_private = True
4404 elif badge_label == 'public':
4405 is_unlisted = is_private = False
4406 return self._availability(is_private, False, False, False, is_unlisted)
4407
4408 @staticmethod
4409 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4410 sidebar_renderer = try_get(
4411 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4412 for item in sidebar_renderer:
4413 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4414 if renderer:
4415 return renderer
4416
4417 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4418 """
4419 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4420 """
4421 browse_id = params = None
4422 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4423 if not renderer:
4424 return
4425 menu_renderer = try_get(
4426 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4427 for menu_item in menu_renderer:
4428 if not isinstance(menu_item, dict):
4429 continue
4430 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4431 text = try_get(
4432 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4433 if not text or text.lower() != 'show unavailable videos':
4434 continue
4435 browse_endpoint = try_get(
4436 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4437 browse_id = browse_endpoint.get('browseId')
4438 params = browse_endpoint.get('params')
4439 break
4440
4441 headers = self.generate_api_headers(
4442 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4443 visitor_data=self._extract_visitor_data(data, ytcfg))
4444 query = {
4445 'params': params or 'wgYCCAA=',
4446 'browseId': browse_id or 'VL%s' % item_id
4447 }
4448 return self._extract_response(
4449 item_id=item_id, headers=headers, query=query,
4450 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4451 note='Downloading API JSON with unavailable videos')
4452
4453 @functools.cached_property
4454 def skip_webpage(self):
4455 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4456
4457 def _extract_webpage(self, url, item_id, fatal=True):
4458 retries = self.get_param('extractor_retries', 3)
4459 count = -1
4460 webpage = data = last_error = None
4461 while count < retries:
4462 count += 1
4463 # Sometimes youtube returns a webpage with incomplete ytInitialData
4464 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4465 if last_error:
4466 self.report_warning('%s. Retrying ...' % last_error)
4467 try:
4468 webpage = self._download_webpage(
4469 url, item_id,
4470 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4471 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4472 except ExtractorError as e:
4473 if isinstance(e.cause, network_exceptions):
4474 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4475 last_error = error_to_compat_str(e.cause or e.msg)
4476 if count < retries:
4477 continue
4478 if fatal:
4479 raise
4480 self.report_warning(error_to_compat_str(e))
4481 break
4482 else:
4483 try:
4484 self._extract_and_report_alerts(data)
4485 except ExtractorError as e:
4486 if fatal:
4487 raise
4488 self.report_warning(error_to_compat_str(e))
4489 break
4490
4491 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4492 break
4493
4494 last_error = 'Incomplete yt initial data received'
4495 if count >= retries:
4496 if fatal:
4497 raise ExtractorError(last_error)
4498 self.report_warning(last_error)
4499 break
4500
4501 return webpage, data
4502
4503 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4504 """Use if failed to extract ytcfg (and data) from initial webpage"""
4505 if not ytcfg and self.is_authenticated:
4506 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4507 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4508 raise ExtractorError(
4509 f'{msg}. If you are not downloading private content, or '
4510 'your cookies are only for the first account and channel,'
4511 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4512 expected=True)
4513 self.report_warning(msg, only_once=True)
4514
4515 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4516 data = None
4517 if not self.skip_webpage:
4518 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4519 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4520 # Reject webpage data if redirected to home page without explicitly requesting
4521 selected_tab = self._extract_selected_tab(traverse_obj(
4522 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4523 if (url != 'https://www.youtube.com/feed/recommended'
4524 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4525 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4526 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4527 if fatal:
4528 raise ExtractorError(msg, expected=True)
4529 self.report_warning(msg, only_once=True)
4530 if not data:
4531 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4532 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4533 return data, ytcfg
4534
4535 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4536 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4537 resolve_response = self._extract_response(
4538 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4539 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4540 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4541 for ep_key, ep in endpoints.items():
4542 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4543 if params:
4544 return self._extract_response(
4545 item_id=item_id, query=params, ep=ep, headers=headers,
4546 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4547 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4548 err_note = 'Failed to resolve url (does the playlist exist?)'
4549 if fatal:
4550 raise ExtractorError(err_note, expected=True)
4551 self.report_warning(err_note, item_id)
4552
4553 _SEARCH_PARAMS = None
4554
4555 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4556 data = {'query': query}
4557 if params is NO_DEFAULT:
4558 params = self._SEARCH_PARAMS
4559 if params:
4560 data['params'] = params
4561
4562 content_keys = (
4563 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4564 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4565 # ytmusic search
4566 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4567 ('continuationContents', ),
4568 )
4569 display_id = f'query "{query}"'
4570 check_get_keys = tuple({keys[0] for keys in content_keys})
4571 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4572 self._report_playlist_authcheck(ytcfg, fatal=False)
4573
4574 continuation_list = [None]
4575 search = None
4576 for page_num in itertools.count(1):
4577 data.update(continuation_list[0] or {})
4578 headers = self.generate_api_headers(
4579 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4580 search = self._extract_response(
4581 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4582 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4583 slr_contents = traverse_obj(search, *content_keys)
4584 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4585 if not continuation_list[0]:
4586 break
4587
4588
4589class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4590 IE_DESC = 'YouTube Tabs'
4591 _VALID_URL = r'''(?x:
4592 https?://
4593 (?:\w+\.)?
4594 (?:
4595 youtube(?:kids)?\.com|
4596 %(invidious)s
4597 )/
4598 (?:
4599 (?P<channel_type>channel|c|user|browse)/|
4600 (?P<not_channel>
4601 feed/|hashtag/|
4602 (?:playlist|watch)\?.*?\blist=
4603 )|
4604 (?!(?:%(reserved_names)s)\b) # Direct URLs
4605 )
4606 (?P<id>[^/?\#&]+)
4607 )''' % {
4608 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4609 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4610 }
4611 IE_NAME = 'youtube:tab'
4612
4613 _TESTS = [{
4614 'note': 'playlists, multipage',
4615 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4616 'playlist_mincount': 94,
4617 'info_dict': {
4618 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4619 'title': 'Igor Kleiner - Playlists',
4620 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4621 'uploader': 'Igor Kleiner',
4622 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4623 'channel': 'Igor Kleiner',
4624 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4625 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4626 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4627 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4628 'channel_follower_count': int
4629 },
4630 }, {
4631 'note': 'playlists, multipage, different order',
4632 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4633 'playlist_mincount': 94,
4634 'info_dict': {
4635 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4636 'title': 'Igor Kleiner - Playlists',
4637 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4638 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4639 'uploader': 'Igor Kleiner',
4640 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4641 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4642 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4643 'channel': 'Igor Kleiner',
4644 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4645 'channel_follower_count': int
4646 },
4647 }, {
4648 'note': 'playlists, series',
4649 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4650 'playlist_mincount': 5,
4651 'info_dict': {
4652 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4653 'title': '3Blue1Brown - Playlists',
4654 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4655 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4656 'uploader': '3Blue1Brown',
4657 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4658 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4659 'channel': '3Blue1Brown',
4660 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4661 'tags': ['Mathematics'],
4662 'channel_follower_count': int
4663 },
4664 }, {
4665 'note': 'playlists, singlepage',
4666 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4667 'playlist_mincount': 4,
4668 'info_dict': {
4669 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4670 'title': 'ThirstForScience - Playlists',
4671 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4672 'uploader': 'ThirstForScience',
4673 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4674 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4675 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4676 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4677 'tags': 'count:13',
4678 'channel': 'ThirstForScience',
4679 'channel_follower_count': int
4680 }
4681 }, {
4682 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4683 'only_matching': True,
4684 }, {
4685 'note': 'basic, single video playlist',
4686 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4687 'info_dict': {
4688 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4689 'uploader': 'Sergey M.',
4690 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4691 'title': 'youtube-dl public playlist',
4692 'description': '',
4693 'tags': [],
4694 'view_count': int,
4695 'modified_date': '20201130',
4696 'channel': 'Sergey M.',
4697 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4698 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4699 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4700 },
4701 'playlist_count': 1,
4702 }, {
4703 'note': 'empty playlist',
4704 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4705 'info_dict': {
4706 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4707 'uploader': 'Sergey M.',
4708 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4709 'title': 'youtube-dl empty playlist',
4710 'tags': [],
4711 'channel': 'Sergey M.',
4712 'description': '',
4713 'modified_date': '20160902',
4714 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4715 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4716 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4717 },
4718 'playlist_count': 0,
4719 }, {
4720 'note': 'Home tab',
4721 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4722 'info_dict': {
4723 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4724 'title': 'lex will - Home',
4725 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4726 'uploader': 'lex will',
4727 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4728 'channel': 'lex will',
4729 'tags': ['bible', 'history', 'prophesy'],
4730 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4731 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4732 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4733 'channel_follower_count': int
4734 },
4735 'playlist_mincount': 2,
4736 }, {
4737 'note': 'Videos tab',
4738 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4739 'info_dict': {
4740 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'title': 'lex will - Videos',
4742 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4743 'uploader': 'lex will',
4744 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4745 'tags': ['bible', 'history', 'prophesy'],
4746 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4747 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4748 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4749 'channel': 'lex will',
4750 'channel_follower_count': int
4751 },
4752 'playlist_mincount': 975,
4753 }, {
4754 'note': 'Videos tab, sorted by popular',
4755 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4756 'info_dict': {
4757 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4758 'title': 'lex will - Videos',
4759 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4760 'uploader': 'lex will',
4761 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4762 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4763 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4764 'channel': 'lex will',
4765 'tags': ['bible', 'history', 'prophesy'],
4766 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4767 'channel_follower_count': int
4768 },
4769 'playlist_mincount': 199,
4770 }, {
4771 'note': 'Playlists tab',
4772 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4773 'info_dict': {
4774 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4775 'title': 'lex will - Playlists',
4776 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4777 'uploader': 'lex will',
4778 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4779 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4780 'channel': 'lex will',
4781 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4782 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4783 'tags': ['bible', 'history', 'prophesy'],
4784 'channel_follower_count': int
4785 },
4786 'playlist_mincount': 17,
4787 }, {
4788 'note': 'Community tab',
4789 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4790 'info_dict': {
4791 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4792 'title': 'lex will - Community',
4793 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4794 'uploader': 'lex will',
4795 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4796 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4797 'channel': 'lex will',
4798 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4799 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4800 'tags': ['bible', 'history', 'prophesy'],
4801 'channel_follower_count': int
4802 },
4803 'playlist_mincount': 18,
4804 }, {
4805 'note': 'Channels tab',
4806 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4807 'info_dict': {
4808 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4809 'title': 'lex will - Channels',
4810 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4811 'uploader': 'lex will',
4812 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4813 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4814 'channel': 'lex will',
4815 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4816 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4817 'tags': ['bible', 'history', 'prophesy'],
4818 'channel_follower_count': int
4819 },
4820 'playlist_mincount': 12,
4821 }, {
4822 'note': 'Search tab',
4823 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4824 'playlist_mincount': 40,
4825 'info_dict': {
4826 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4827 'title': '3Blue1Brown - Search - linear algebra',
4828 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4829 'uploader': '3Blue1Brown',
4830 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4831 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4832 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4833 'tags': ['Mathematics'],
4834 'channel': '3Blue1Brown',
4835 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4836 'channel_follower_count': int
4837 },
4838 }, {
4839 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4840 'only_matching': True,
4841 }, {
4842 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4843 'only_matching': True,
4844 }, {
4845 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4846 'only_matching': True,
4847 }, {
4848 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4849 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4850 'info_dict': {
4851 'title': '29C3: Not my department',
4852 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4853 'uploader': 'Christiaan008',
4854 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4855 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4856 'tags': [],
4857 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4858 'view_count': int,
4859 'modified_date': '20150605',
4860 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4861 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4862 'channel': 'Christiaan008',
4863 },
4864 'playlist_count': 96,
4865 }, {
4866 'note': 'Large playlist',
4867 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4868 'info_dict': {
4869 'title': 'Uploads from Cauchemar',
4870 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4871 'uploader': 'Cauchemar',
4872 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4873 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4874 'tags': [],
4875 'modified_date': r're:\d{8}',
4876 'channel': 'Cauchemar',
4877 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4878 'view_count': int,
4879 'description': '',
4880 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4881 },
4882 'playlist_mincount': 1123,
4883 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4884 }, {
4885 'note': 'even larger playlist, 8832 videos',
4886 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4887 'only_matching': True,
4888 }, {
4889 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4890 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4891 'info_dict': {
4892 'title': 'Uploads from Interstellar Movie',
4893 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4894 'uploader': 'Interstellar Movie',
4895 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4896 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4897 'tags': [],
4898 'view_count': int,
4899 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4900 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4901 'channel': 'Interstellar Movie',
4902 'description': '',
4903 'modified_date': r're:\d{8}',
4904 },
4905 'playlist_mincount': 21,
4906 }, {
4907 'note': 'Playlist with "show unavailable videos" button',
4908 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4909 'info_dict': {
4910 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4911 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4912 'uploader': 'Phim Siêu Nhân Nhật Bản',
4913 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4914 'view_count': int,
4915 'channel': 'Phim Siêu Nhân Nhật Bản',
4916 'tags': [],
4917 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4918 'description': '',
4919 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4920 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4921 'modified_date': r're:\d{8}',
4922 },
4923 'playlist_mincount': 200,
4924 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4925 }, {
4926 'note': 'Playlist with unavailable videos in page 7',
4927 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4928 'info_dict': {
4929 'title': 'Uploads from BlankTV',
4930 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4931 'uploader': 'BlankTV',
4932 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4933 'channel': 'BlankTV',
4934 'channel_url': 'https://www.youtube.com/c/blanktv',
4935 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4936 'view_count': int,
4937 'tags': [],
4938 'uploader_url': 'https://www.youtube.com/c/blanktv',
4939 'modified_date': r're:\d{8}',
4940 'description': '',
4941 },
4942 'playlist_mincount': 1000,
4943 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4944 }, {
4945 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4946 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4947 'info_dict': {
4948 'title': 'Data Analysis with Dr Mike Pound',
4949 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4950 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4951 'uploader': 'Computerphile',
4952 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4953 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4954 'tags': [],
4955 'view_count': int,
4956 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4957 'channel_url': 'https://www.youtube.com/user/Computerphile',
4958 'channel': 'Computerphile',
4959 },
4960 'playlist_mincount': 11,
4961 }, {
4962 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4963 'only_matching': True,
4964 }, {
4965 'note': 'Playlist URL that does not actually serve a playlist',
4966 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4967 'info_dict': {
4968 'id': 'FqZTN594JQw',
4969 'ext': 'webm',
4970 'title': "Smiley's People 01 detective, Adventure Series, Action",
4971 'uploader': 'STREEM',
4972 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4973 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4974 'upload_date': '20150526',
4975 'license': 'Standard YouTube License',
4976 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4977 'categories': ['People & Blogs'],
4978 'tags': list,
4979 'view_count': int,
4980 'like_count': int,
4981 },
4982 'params': {
4983 'skip_download': True,
4984 },
4985 'skip': 'This video is not available.',
4986 'add_ie': [YoutubeIE.ie_key()],
4987 }, {
4988 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4989 'only_matching': True,
4990 }, {
4991 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4992 'only_matching': True,
4993 }, {
4994 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4995 'info_dict': {
4996 'id': 'GgL890LIznQ', # This will keep changing
4997 'ext': 'mp4',
4998 'title': str,
4999 'uploader': 'Sky News',
5000 'uploader_id': 'skynews',
5001 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
5002 'upload_date': r're:\d{8}',
5003 'description': str,
5004 'categories': ['News & Politics'],
5005 'tags': list,
5006 'like_count': int,
5007 'release_timestamp': 1642502819,
5008 'channel': 'Sky News',
5009 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5010 'age_limit': 0,
5011 'view_count': int,
5012 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
5013 'playable_in_embed': True,
5014 'release_date': '20220118',
5015 'availability': 'public',
5016 'live_status': 'is_live',
5017 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5018 'channel_follower_count': int
5019 },
5020 'params': {
5021 'skip_download': True,
5022 },
5023 'expected_warnings': ['Ignoring subtitle tracks found in '],
5024 }, {
5025 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5026 'info_dict': {
5027 'id': 'a48o2S1cPoo',
5028 'ext': 'mp4',
5029 'title': 'The Young Turks - Live Main Show',
5030 'uploader': 'The Young Turks',
5031 'uploader_id': 'TheYoungTurks',
5032 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
5033 'upload_date': '20150715',
5034 'license': 'Standard YouTube License',
5035 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5036 'categories': ['News & Politics'],
5037 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5038 'like_count': int,
5039 },
5040 'params': {
5041 'skip_download': True,
5042 },
5043 'only_matching': True,
5044 }, {
5045 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5046 'only_matching': True,
5047 }, {
5048 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5049 'only_matching': True,
5050 }, {
5051 'note': 'A channel that is not live. Should raise error',
5052 'url': 'https://www.youtube.com/user/numberphile/live',
5053 'only_matching': True,
5054 }, {
5055 'url': 'https://www.youtube.com/feed/trending',
5056 'only_matching': True,
5057 }, {
5058 'url': 'https://www.youtube.com/feed/library',
5059 'only_matching': True,
5060 }, {
5061 'url': 'https://www.youtube.com/feed/history',
5062 'only_matching': True,
5063 }, {
5064 'url': 'https://www.youtube.com/feed/subscriptions',
5065 'only_matching': True,
5066 }, {
5067 'url': 'https://www.youtube.com/feed/watch_later',
5068 'only_matching': True,
5069 }, {
5070 'note': 'Recommended - redirects to home page.',
5071 'url': 'https://www.youtube.com/feed/recommended',
5072 'only_matching': True,
5073 }, {
5074 'note': 'inline playlist with not always working continuations',
5075 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5076 'only_matching': True,
5077 }, {
5078 'url': 'https://www.youtube.com/course',
5079 'only_matching': True,
5080 }, {
5081 'url': 'https://www.youtube.com/zsecurity',
5082 'only_matching': True,
5083 }, {
5084 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5085 'only_matching': True,
5086 }, {
5087 'url': 'https://www.youtube.com/TheYoungTurks/live',
5088 'only_matching': True,
5089 }, {
5090 'url': 'https://www.youtube.com/hashtag/cctv9',
5091 'info_dict': {
5092 'id': 'cctv9',
5093 'title': '#cctv9',
5094 'tags': [],
5095 },
5096 'playlist_mincount': 350,
5097 }, {
5098 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5099 'only_matching': True,
5100 }, {
5101 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5102 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5103 'only_matching': True
5104 }, {
5105 'note': '/browse/ should redirect to /channel/',
5106 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5107 'only_matching': True
5108 }, {
5109 'note': 'VLPL, should redirect to playlist?list=PL...',
5110 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5111 'info_dict': {
5112 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5113 'uploader': 'NoCopyrightSounds',
5114 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5115 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5116 'title': 'NCS Releases',
5117 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5118 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5119 'modified_date': r're:\d{8}',
5120 'view_count': int,
5121 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5122 'tags': [],
5123 'channel': 'NoCopyrightSounds',
5124 },
5125 'playlist_mincount': 166,
5126 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5127 }, {
5128 'note': 'Topic, should redirect to playlist?list=UU...',
5129 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5130 'info_dict': {
5131 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5132 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5133 'title': 'Uploads from Royalty Free Music - Topic',
5134 'uploader': 'Royalty Free Music - Topic',
5135 'tags': [],
5136 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5137 'channel': 'Royalty Free Music - Topic',
5138 'view_count': int,
5139 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5140 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5141 'modified_date': r're:\d{8}',
5142 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5143 'description': '',
5144 },
5145 'expected_warnings': [
5146 'The URL does not have a videos tab',
5147 r'[Uu]navailable videos (are|will be) hidden',
5148 ],
5149 'playlist_mincount': 101,
5150 }, {
5151 'note': 'Topic without a UU playlist',
5152 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5153 'info_dict': {
5154 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5155 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5156 'tags': [],
5157 },
5158 'expected_warnings': [
5159 'the playlist redirect gave error',
5160 ],
5161 'playlist_mincount': 9,
5162 }, {
5163 'note': 'Youtube music Album',
5164 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5165 'info_dict': {
5166 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5167 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5168 'tags': [],
5169 'view_count': int,
5170 'description': '',
5171 'availability': 'unlisted',
5172 'modified_date': r're:\d{8}',
5173 },
5174 'playlist_count': 50,
5175 }, {
5176 'note': 'unlisted single video playlist',
5177 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5178 'info_dict': {
5179 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5180 'uploader': 'colethedj',
5181 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5182 'title': 'yt-dlp unlisted playlist test',
5183 'availability': 'unlisted',
5184 'tags': [],
5185 'modified_date': '20211208',
5186 'channel': 'colethedj',
5187 'view_count': int,
5188 'description': '',
5189 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5190 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5191 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5192 },
5193 'playlist_count': 1,
5194 }, {
5195 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5196 'url': 'https://www.youtube.com/feed/recommended',
5197 'info_dict': {
5198 'id': 'recommended',
5199 'title': 'recommended',
5200 'tags': [],
5201 },
5202 'playlist_mincount': 50,
5203 'params': {
5204 'skip_download': True,
5205 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5206 },
5207 }, {
5208 'note': 'API Fallback: /videos tab, sorted by oldest first',
5209 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5210 'info_dict': {
5211 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5212 'title': 'Cody\'sLab - Videos',
5213 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5214 'uploader': 'Cody\'sLab',
5215 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5216 'channel': 'Cody\'sLab',
5217 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5218 'tags': [],
5219 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5220 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5221 'channel_follower_count': int
5222 },
5223 'playlist_mincount': 650,
5224 'params': {
5225 'skip_download': True,
5226 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5227 },
5228 }, {
5229 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5230 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5231 'info_dict': {
5232 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5233 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5234 'title': 'Uploads from Royalty Free Music - Topic',
5235 'uploader': 'Royalty Free Music - Topic',
5236 'modified_date': r're:\d{8}',
5237 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5238 'description': '',
5239 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5240 'tags': [],
5241 'channel': 'Royalty Free Music - Topic',
5242 'view_count': int,
5243 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5244 },
5245 'expected_warnings': [
5246 'does not have a videos tab',
5247 r'[Uu]navailable videos (are|will be) hidden',
5248 ],
5249 'playlist_mincount': 101,
5250 'params': {
5251 'skip_download': True,
5252 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5253 },
5254 }, {
5255 'note': 'non-standard redirect to regional channel',
5256 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5257 'only_matching': True
5258 }, {
5259 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5260 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5261 'info_dict': {
5262 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5263 'modified_date': '20220407',
5264 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5265 'tags': [],
5266 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5267 'uploader': 'pukkandan',
5268 'availability': 'unlisted',
5269 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5270 'channel': 'pukkandan',
5271 'description': 'Test for collaborative playlist',
5272 'title': 'yt-dlp test - collaborative playlist',
5273 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5274 },
5275 'playlist_mincount': 2
5276 }]
5277
5278 @classmethod
5279 def suitable(cls, url):
5280 return False if YoutubeIE.suitable(url) else super().suitable(url)
5281
5282 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5283
5284 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5285 def _real_extract(self, url, smuggled_data):
5286 item_id = self._match_id(url)
5287 url = compat_urlparse.urlunparse(
5288 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5289 compat_opts = self.get_param('compat_opts', [])
5290
5291 def get_mobj(url):
5292 mobj = self._URL_RE.match(url).groupdict()
5293 mobj.update((k, '') for k, v in mobj.items() if v is None)
5294 return mobj
5295
5296 mobj, redirect_warning = get_mobj(url), None
5297 # Youtube returns incomplete data if tabname is not lower case
5298 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5299 if is_channel:
5300 if smuggled_data.get('is_music_url'):
5301 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5302 item_id = item_id[2:]
5303 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5304 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5305 mdata = self._extract_tab_endpoint(
5306 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5307 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5308 get_all=False, expected_type=compat_str)
5309 if not murl:
5310 raise ExtractorError('Failed to resolve album to playlist')
5311 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5312 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5313 pre = f'https://www.youtube.com/channel/{item_id}'
5314
5315 original_tab_name = tab
5316 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5317 # Home URLs should redirect to /videos/
5318 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5319 'To download only the videos in the home page, add a "/featured" to the URL')
5320 tab = '/videos'
5321
5322 url = ''.join((pre, tab, post))
5323 mobj = get_mobj(url)
5324
5325 # Handle both video/playlist URLs
5326 qs = parse_qs(url)
5327 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5328
5329 if not video_id and mobj['not_channel'].startswith('watch'):
5330 if not playlist_id:
5331 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5332 raise ExtractorError('Unable to recognize tab page')
5333 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5334 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5335 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5336 mobj = get_mobj(url)
5337
5338 if video_id and playlist_id:
5339 if self.get_param('noplaylist'):
5340 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5341 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5342 ie=YoutubeIE.ie_key(), video_id=video_id)
5343 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5344
5345 data, ytcfg = self._extract_data(url, item_id)
5346
5347 # YouTube may provide a non-standard redirect to the regional channel
5348 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5349 redirect_url = traverse_obj(
5350 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5351 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5352 redirect_url = ''.join((
5353 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5354 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5355 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5356
5357 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5358 if tabs:
5359 selected_tab = self._extract_selected_tab(tabs)
5360 selected_tab_name = selected_tab.get('title', '').lower()
5361 if selected_tab_name == 'home':
5362 selected_tab_name = 'featured'
5363 requested_tab_name = mobj['tab'][1:]
5364 if 'no-youtube-channel-redirect' not in compat_opts:
5365 if requested_tab_name == 'live':
5366 # Live tab should have redirected to the video
5367 raise ExtractorError('The channel is not currently live', expected=True)
5368 if requested_tab_name not in ('', selected_tab_name):
5369 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5370 if not original_tab_name:
5371 if item_id[:2] == 'UC':
5372 # Topic channels don't have /videos. Use the equivalent playlist instead
5373 pl_id = f'UU{item_id[2:]}'
5374 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5375 try:
5376 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5377 except ExtractorError:
5378 redirect_warning += ' and the playlist redirect gave error'
5379 else:
5380 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5381 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5382 if selected_tab_name and selected_tab_name != requested_tab_name:
5383 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5384 else:
5385 raise ExtractorError(redirect_warning, expected=True)
5386
5387 if redirect_warning:
5388 self.to_screen(redirect_warning)
5389 self.write_debug(f'Final URL: {url}')
5390
5391 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5392 if 'no-youtube-unavailable-videos' not in compat_opts:
5393 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5394 self._extract_and_report_alerts(data, only_once=True)
5395 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5396 if tabs:
5397 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5398
5399 playlist = traverse_obj(
5400 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5401 if playlist:
5402 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5403
5404 video_id = traverse_obj(
5405 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5406 if video_id:
5407 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5408 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5409 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5410 ie=YoutubeIE.ie_key(), video_id=video_id)
5411
5412 raise ExtractorError('Unable to recognize tab page')
5413
5414
5415class YoutubePlaylistIE(InfoExtractor):
5416 IE_DESC = 'YouTube playlists'
5417 _VALID_URL = r'''(?x)(?:
5418 (?:https?://)?
5419 (?:\w+\.)?
5420 (?:
5421 (?:
5422 youtube(?:kids)?\.com|
5423 %(invidious)s
5424 )
5425 /.*?\?.*?\blist=
5426 )?
5427 (?P<id>%(playlist_id)s)
5428 )''' % {
5429 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5430 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5431 }
5432 IE_NAME = 'youtube:playlist'
5433 _TESTS = [{
5434 'note': 'issue #673',
5435 'url': 'PLBB231211A4F62143',
5436 'info_dict': {
5437 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5438 'id': 'PLBB231211A4F62143',
5439 'uploader': 'Wickman',
5440 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5441 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5442 'view_count': int,
5443 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5444 'modified_date': r're:\d{8}',
5445 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5446 'channel': 'Wickman',
5447 'tags': [],
5448 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5449 },
5450 'playlist_mincount': 29,
5451 }, {
5452 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5453 'info_dict': {
5454 'title': 'YDL_safe_search',
5455 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5456 },
5457 'playlist_count': 2,
5458 'skip': 'This playlist is private',
5459 }, {
5460 'note': 'embedded',
5461 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5462 'playlist_count': 4,
5463 'info_dict': {
5464 'title': 'JODA15',
5465 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5466 'uploader': 'milan',
5467 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5468 'description': '',
5469 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5470 'tags': [],
5471 'modified_date': '20140919',
5472 'view_count': int,
5473 'channel': 'milan',
5474 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5475 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5476 },
5477 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5478 }, {
5479 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5480 'playlist_mincount': 654,
5481 'info_dict': {
5482 'title': '2018 Chinese New Singles (11/6 updated)',
5483 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5484 'uploader': 'LBK',
5485 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5486 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5487 'channel': 'LBK',
5488 'view_count': int,
5489 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5490 'tags': [],
5491 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5492 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5493 'modified_date': r're:\d{8}',
5494 },
5495 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5496 }, {
5497 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5498 'only_matching': True,
5499 }, {
5500 # music album playlist
5501 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5502 'only_matching': True,
5503 }]
5504
5505 @classmethod
5506 def suitable(cls, url):
5507 if YoutubeTabIE.suitable(url):
5508 return False
5509 from ..utils import parse_qs
5510 qs = parse_qs(url)
5511 if qs.get('v', [None])[0]:
5512 return False
5513 return super().suitable(url)
5514
5515 def _real_extract(self, url):
5516 playlist_id = self._match_id(url)
5517 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5518 url = update_url_query(
5519 'https://www.youtube.com/playlist',
5520 parse_qs(url) or {'list': playlist_id})
5521 if is_music_url:
5522 url = smuggle_url(url, {'is_music_url': True})
5523 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5524
5525
5526class YoutubeYtBeIE(InfoExtractor):
5527 IE_DESC = 'youtu.be'
5528 _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}
5529 _TESTS = [{
5530 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5531 'info_dict': {
5532 'id': 'yeWKywCrFtk',
5533 'ext': 'mp4',
5534 'title': 'Small Scale Baler and Braiding Rugs',
5535 'uploader': 'Backus-Page House Museum',
5536 'uploader_id': 'backuspagemuseum',
5537 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5538 'upload_date': '20161008',
5539 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5540 'categories': ['Nonprofits & Activism'],
5541 'tags': list,
5542 'like_count': int,
5543 'age_limit': 0,
5544 'playable_in_embed': True,
5545 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5546 'channel': 'Backus-Page House Museum',
5547 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5548 'live_status': 'not_live',
5549 'view_count': int,
5550 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5551 'availability': 'public',
5552 'duration': 59,
5553 },
5554 'params': {
5555 'noplaylist': True,
5556 'skip_download': True,
5557 },
5558 }, {
5559 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5560 'only_matching': True,
5561 }]
5562
5563 def _real_extract(self, url):
5564 mobj = self._match_valid_url(url)
5565 video_id = mobj.group('id')
5566 playlist_id = mobj.group('playlist_id')
5567 return self.url_result(
5568 update_url_query('https://www.youtube.com/watch', {
5569 'v': video_id,
5570 'list': playlist_id,
5571 'feature': 'youtu.be',
5572 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5573
5574
5575class YoutubeLivestreamEmbedIE(InfoExtractor):
5576 IE_DESC = 'YouTube livestream embeds'
5577 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5578 _TESTS = [{
5579 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5580 'only_matching': True,
5581 }]
5582
5583 def _real_extract(self, url):
5584 channel_id = self._match_id(url)
5585 return self.url_result(
5586 f'https://www.youtube.com/channel/{channel_id}/live',
5587 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5588
5589
5590class YoutubeYtUserIE(InfoExtractor):
5591 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5592 IE_NAME = 'youtube:user'
5593 _VALID_URL = r'ytuser:(?P<id>.+)'
5594 _TESTS = [{
5595 'url': 'ytuser:phihag',
5596 'only_matching': True,
5597 }]
5598
5599 def _real_extract(self, url):
5600 user_id = self._match_id(url)
5601 return self.url_result(
5602 'https://www.youtube.com/user/%s/videos' % user_id,
5603 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5604
5605
5606class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5607 IE_NAME = 'youtube:favorites'
5608 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5609 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5610 _LOGIN_REQUIRED = True
5611 _TESTS = [{
5612 'url': ':ytfav',
5613 'only_matching': True,
5614 }, {
5615 'url': ':ytfavorites',
5616 'only_matching': True,
5617 }]
5618
5619 def _real_extract(self, url):
5620 return self.url_result(
5621 'https://www.youtube.com/playlist?list=LL',
5622 ie=YoutubeTabIE.ie_key())
5623
5624
5625class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5626 IE_NAME = 'youtube:notif'
5627 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5628 _VALID_URL = r':ytnotif(?:ication)?s?'
5629 _LOGIN_REQUIRED = True
5630 _TESTS = [{
5631 'url': ':ytnotif',
5632 'only_matching': True,
5633 }, {
5634 'url': ':ytnotifications',
5635 'only_matching': True,
5636 }]
5637
5638 def _extract_notification_menu(self, response, continuation_list):
5639 notification_list = traverse_obj(
5640 response,
5641 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5642 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5643 expected_type=list) or []
5644 continuation_list[0] = None
5645 for item in notification_list:
5646 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5647 if entry:
5648 yield entry
5649 continuation = item.get('continuationItemRenderer')
5650 if continuation:
5651 continuation_list[0] = continuation
5652
5653 def _extract_notification_renderer(self, notification):
5654 video_id = traverse_obj(
5655 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5656 url = f'https://www.youtube.com/watch?v={video_id}'
5657 channel_id = None
5658 if not video_id:
5659 browse_ep = traverse_obj(
5660 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5661 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5662 post_id = self._search_regex(
5663 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5664 'post id', default=None)
5665 if not channel_id or not post_id:
5666 return
5667 # The direct /post url redirects to this in the browser
5668 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5669
5670 channel = traverse_obj(
5671 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5672 expected_type=str)
5673 notification_title = self._get_text(notification, 'shortMessage')
5674 if notification_title:
5675 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
5676 # TODO: handle recommended videos
5677 title = self._search_regex(
5678 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
5679 'video title', default=None)
5680 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5681 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5682 else None)
5683 return {
5684 '_type': 'url',
5685 'url': url,
5686 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5687 'video_id': video_id,
5688 'title': title,
5689 'channel_id': channel_id,
5690 'channel': channel,
5691 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5692 'upload_date': upload_date,
5693 }
5694
5695 def _notification_menu_entries(self, ytcfg):
5696 continuation_list = [None]
5697 response = None
5698 for page in itertools.count(1):
5699 ctoken = traverse_obj(
5700 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5701 response = self._extract_response(
5702 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5703 ep='notification/get_notification_menu', check_get_keys='actions',
5704 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5705 yield from self._extract_notification_menu(response, continuation_list)
5706 if not continuation_list[0]:
5707 break
5708
5709 def _real_extract(self, url):
5710 display_id = 'notifications'
5711 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5712 self._report_playlist_authcheck(ytcfg)
5713 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5714
5715
5716class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5717 IE_DESC = 'YouTube search'
5718 IE_NAME = 'youtube:search'
5719 _SEARCH_KEY = 'ytsearch'
5720 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5721 _TESTS = [{
5722 'url': 'ytsearch5:youtube-dl test video',
5723 'playlist_count': 5,
5724 'info_dict': {
5725 'id': 'youtube-dl test video',
5726 'title': 'youtube-dl test video',
5727 }
5728 }]
5729
5730
5731class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5732 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5733 _SEARCH_KEY = 'ytsearchdate'
5734 IE_DESC = 'YouTube search, newest videos first'
5735 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5736 _TESTS = [{
5737 'url': 'ytsearchdate5:youtube-dl test video',
5738 'playlist_count': 5,
5739 'info_dict': {
5740 'id': 'youtube-dl test video',
5741 'title': 'youtube-dl test video',
5742 }
5743 }]
5744
5745
5746class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5747 IE_DESC = 'YouTube search URLs with sorting and filter support'
5748 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5749 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5750 _TESTS = [{
5751 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5752 'playlist_mincount': 5,
5753 'info_dict': {
5754 'id': 'youtube-dl test video',
5755 'title': 'youtube-dl test video',
5756 }
5757 }, {
5758 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5759 'playlist_mincount': 5,
5760 'info_dict': {
5761 'id': 'python',
5762 'title': 'python',
5763 }
5764 }, {
5765 'url': 'https://www.youtube.com/results?search_query=%23cats',
5766 'playlist_mincount': 1,
5767 'info_dict': {
5768 'id': '#cats',
5769 'title': '#cats',
5770 'entries': [{
5771 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5772 'title': '#cats',
5773 }],
5774 },
5775 }, {
5776 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5777 'only_matching': True,
5778 }]
5779
5780 def _real_extract(self, url):
5781 qs = parse_qs(url)
5782 query = (qs.get('search_query') or qs.get('q'))[0]
5783 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5784
5785
5786class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5787 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5788 IE_NAME = 'youtube:music:search_url'
5789 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5790 _TESTS = [{
5791 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5792 'playlist_count': 16,
5793 'info_dict': {
5794 'id': 'royalty free music',
5795 'title': 'royalty free music',
5796 }
5797 }, {
5798 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5799 'playlist_mincount': 30,
5800 'info_dict': {
5801 'id': 'royalty free music - songs',
5802 'title': 'royalty free music - songs',
5803 },
5804 'params': {'extract_flat': 'in_playlist'}
5805 }, {
5806 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5807 'playlist_mincount': 30,
5808 'info_dict': {
5809 'id': 'royalty free music - community playlists',
5810 'title': 'royalty free music - community playlists',
5811 },
5812 'params': {'extract_flat': 'in_playlist'}
5813 }]
5814
5815 _SECTIONS = {
5816 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5817 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5818 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5819 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5820 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5821 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5822 }
5823
5824 def _real_extract(self, url):
5825 qs = parse_qs(url)
5826 query = (qs.get('search_query') or qs.get('q'))[0]
5827 params = qs.get('sp', (None,))[0]
5828 if params:
5829 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5830 else:
5831 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5832 params = self._SECTIONS.get(section)
5833 if not params:
5834 section = None
5835 title = join_nonempty(query, section, delim=' - ')
5836 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5837
5838
5839class YoutubeFeedsInfoExtractor(InfoExtractor):
5840 """
5841 Base class for feed extractors
5842 Subclasses must re-define the _FEED_NAME property.
5843 """
5844 _LOGIN_REQUIRED = True
5845 _FEED_NAME = 'feeds'
5846
5847 def _real_initialize(self):
5848 YoutubeBaseInfoExtractor._check_login_required(self)
5849
5850 @classproperty
5851 def IE_NAME(self):
5852 return f'youtube:{self._FEED_NAME}'
5853
5854 def _real_extract(self, url):
5855 return self.url_result(
5856 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5857
5858
5859class YoutubeWatchLaterIE(InfoExtractor):
5860 IE_NAME = 'youtube:watchlater'
5861 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5862 _VALID_URL = r':ytwatchlater'
5863 _TESTS = [{
5864 'url': ':ytwatchlater',
5865 'only_matching': True,
5866 }]
5867
5868 def _real_extract(self, url):
5869 return self.url_result(
5870 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5871
5872
5873class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5874 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5875 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5876 _FEED_NAME = 'recommended'
5877 _LOGIN_REQUIRED = False
5878 _TESTS = [{
5879 'url': ':ytrec',
5880 'only_matching': True,
5881 }, {
5882 'url': ':ytrecommended',
5883 'only_matching': True,
5884 }, {
5885 'url': 'https://youtube.com',
5886 'only_matching': True,
5887 }]
5888
5889
5890class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5891 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5892 _VALID_URL = r':ytsub(?:scription)?s?'
5893 _FEED_NAME = 'subscriptions'
5894 _TESTS = [{
5895 'url': ':ytsubs',
5896 'only_matching': True,
5897 }, {
5898 'url': ':ytsubscriptions',
5899 'only_matching': True,
5900 }]
5901
5902
5903class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5904 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5905 _VALID_URL = r':ythis(?:tory)?'
5906 _FEED_NAME = 'history'
5907 _TESTS = [{
5908 'url': ':ythistory',
5909 'only_matching': True,
5910 }]
5911
5912
5913class YoutubeStoriesIE(InfoExtractor):
5914 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5915 IE_NAME = 'youtube:stories'
5916 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5917 _TESTS = [{
5918 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5919 'only_matching': True,
5920 }]
5921
5922 def _real_extract(self, url):
5923 playlist_id = f'RLTD{self._match_id(url)}'
5924 return self.url_result(
5925 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5926 ie=YoutubeTabIE, video_id=playlist_id)
5927
5928
5929class YoutubeTruncatedURLIE(InfoExtractor):
5930 IE_NAME = 'youtube:truncated_url'
5931 IE_DESC = False # Do not list
5932 _VALID_URL = r'''(?x)
5933 (?:https?://)?
5934 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5935 (?:watch\?(?:
5936 feature=[a-z_]+|
5937 annotation_id=annotation_[^&]+|
5938 x-yt-cl=[0-9]+|
5939 hl=[^&]*|
5940 t=[0-9]+
5941 )?
5942 |
5943 attribution_link\?a=[^&]+
5944 )
5945 $
5946 '''
5947
5948 _TESTS = [{
5949 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5950 'only_matching': True,
5951 }, {
5952 'url': 'https://www.youtube.com/watch?',
5953 'only_matching': True,
5954 }, {
5955 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5956 'only_matching': True,
5957 }, {
5958 'url': 'https://www.youtube.com/watch?feature=foo',
5959 'only_matching': True,
5960 }, {
5961 'url': 'https://www.youtube.com/watch?hl=en-GB',
5962 'only_matching': True,
5963 }, {
5964 'url': 'https://www.youtube.com/watch?t=2372',
5965 'only_matching': True,
5966 }]
5967
5968 def _real_extract(self, url):
5969 raise ExtractorError(
5970 'Did you forget to quote the URL? Remember that & is a meta '
5971 'character in most shells, so you want to put the URL in quotes, '
5972 'like youtube-dl '
5973 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5974 ' or simply youtube-dl BaW_jenozKc .',
5975 expected=True)
5976
5977
5978class YoutubeClipIE(InfoExtractor):
5979 IE_NAME = 'youtube:clip'
5980 IE_DESC = False # Do not list
5981 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5982
5983 def _real_extract(self, url):
5984 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5985 return self.url_result(url, 'Generic')
5986
5987
5988class YoutubeTruncatedIDIE(InfoExtractor):
5989 IE_NAME = 'youtube:truncated_id'
5990 IE_DESC = False # Do not list
5991 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5992
5993 _TESTS = [{
5994 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5995 'only_matching': True,
5996 }]
5997
5998 def _real_extract(self, url):
5999 video_id = self._match_id(url)
6000 raise ExtractorError(
6001 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
6002 expected=True)