]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[ruutu] Detect embeds (#3294)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
... / ...
CommitLineData
1# coding: utf-8
2
3from __future__ import unicode_literals
4
5import calendar
6import copy
7import datetime
8import functools
9import hashlib
10import itertools
11import json
12import math
13import os.path
14import random
15import re
16import sys
17import time
18import traceback
19import threading
20
21from .common import InfoExtractor, SearchInfoExtractor
22from ..compat import (
23 compat_chr,
24 compat_HTTPError,
25 compat_parse_qs,
26 compat_str,
27 compat_urllib_parse_unquote_plus,
28 compat_urllib_parse_urlencode,
29 compat_urllib_parse_urlparse,
30 compat_urlparse,
31)
32from ..jsinterp import JSInterpreter
33from ..utils import (
34 bug_reports_message,
35 clean_html,
36 datetime_from_str,
37 dict_get,
38 error_to_compat_str,
39 ExtractorError,
40 float_or_none,
41 format_field,
42 get_first,
43 int_or_none,
44 is_html,
45 join_nonempty,
46 js_to_json,
47 mimetype2ext,
48 network_exceptions,
49 NO_DEFAULT,
50 orderedSet,
51 parse_codecs,
52 parse_count,
53 parse_duration,
54 parse_iso8601,
55 parse_qs,
56 qualities,
57 remove_end,
58 remove_start,
59 smuggle_url,
60 str_or_none,
61 str_to_int,
62 strftime_or_none,
63 traverse_obj,
64 try_get,
65 unescapeHTML,
66 unified_strdate,
67 unified_timestamp,
68 unsmuggle_url,
69 update_url_query,
70 url_or_none,
71 urljoin,
72 variadic,
73)
74
75
76# any clients starting with _ cannot be explicity requested by the user
77INNERTUBE_CLIENTS = {
78 'web': {
79 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
80 'INNERTUBE_CONTEXT': {
81 'client': {
82 'clientName': 'WEB',
83 'clientVersion': '2.20211221.00.00',
84 }
85 },
86 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
87 },
88 'web_embedded': {
89 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
90 'INNERTUBE_CONTEXT': {
91 'client': {
92 'clientName': 'WEB_EMBEDDED_PLAYER',
93 'clientVersion': '1.20211215.00.01',
94 },
95 },
96 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
97 },
98 'web_music': {
99 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
100 'INNERTUBE_HOST': 'music.youtube.com',
101 'INNERTUBE_CONTEXT': {
102 'client': {
103 'clientName': 'WEB_REMIX',
104 'clientVersion': '1.20211213.00.00',
105 }
106 },
107 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
108 },
109 'web_creator': {
110 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
111 'INNERTUBE_CONTEXT': {
112 'client': {
113 'clientName': 'WEB_CREATOR',
114 'clientVersion': '1.20211220.02.00',
115 }
116 },
117 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
118 },
119 'android': {
120 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
121 'INNERTUBE_CONTEXT': {
122 'client': {
123 'clientName': 'ANDROID',
124 'clientVersion': '16.49',
125 }
126 },
127 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
128 'REQUIRE_JS_PLAYER': False
129 },
130 'android_embedded': {
131 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
132 'INNERTUBE_CONTEXT': {
133 'client': {
134 'clientName': 'ANDROID_EMBEDDED_PLAYER',
135 'clientVersion': '16.49',
136 },
137 },
138 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
139 'REQUIRE_JS_PLAYER': False
140 },
141 'android_music': {
142 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
143 'INNERTUBE_CONTEXT': {
144 'client': {
145 'clientName': 'ANDROID_MUSIC',
146 'clientVersion': '4.57',
147 }
148 },
149 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
150 'REQUIRE_JS_PLAYER': False
151 },
152 'android_creator': {
153 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
154 'INNERTUBE_CONTEXT': {
155 'client': {
156 'clientName': 'ANDROID_CREATOR',
157 'clientVersion': '21.47',
158 },
159 },
160 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
161 'REQUIRE_JS_PLAYER': False
162 },
163 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
164 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
165 'ios': {
166 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
167 'INNERTUBE_CONTEXT': {
168 'client': {
169 'clientName': 'IOS',
170 'clientVersion': '16.46',
171 'deviceModel': 'iPhone14,3',
172 }
173 },
174 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
175 'REQUIRE_JS_PLAYER': False
176 },
177 'ios_embedded': {
178 'INNERTUBE_CONTEXT': {
179 'client': {
180 'clientName': 'IOS_MESSAGES_EXTENSION',
181 'clientVersion': '16.46',
182 'deviceModel': 'iPhone14,3',
183 },
184 },
185 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
186 'REQUIRE_JS_PLAYER': False
187 },
188 'ios_music': {
189 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
190 'INNERTUBE_CONTEXT': {
191 'client': {
192 'clientName': 'IOS_MUSIC',
193 'clientVersion': '4.57',
194 },
195 },
196 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
197 'REQUIRE_JS_PLAYER': False
198 },
199 'ios_creator': {
200 'INNERTUBE_CONTEXT': {
201 'client': {
202 'clientName': 'IOS_CREATOR',
203 'clientVersion': '21.47',
204 },
205 },
206 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
207 'REQUIRE_JS_PLAYER': False
208 },
209 # mweb has 'ultralow' formats
210 # See: https://github.com/yt-dlp/yt-dlp/pull/557
211 'mweb': {
212 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
213 'INNERTUBE_CONTEXT': {
214 'client': {
215 'clientName': 'MWEB',
216 'clientVersion': '2.20211221.01.00',
217 }
218 },
219 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
220 },
221 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
222 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
223 'tv_embedded': {
224 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
225 'INNERTUBE_CONTEXT': {
226 'client': {
227 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
228 'clientVersion': '2.0',
229 },
230 },
231 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
232 },
233}
234
235
236def _split_innertube_client(client_name):
237 variant, *base = client_name.rsplit('.', 1)
238 if base:
239 return variant, base[0], variant
240 base, *variant = client_name.split('_', 1)
241 return client_name, base, variant[0] if variant else None
242
243
244def build_innertube_clients():
245 THIRD_PARTY = {
246 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
247 }
248 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
249 priority = qualities(BASE_CLIENTS[::-1])
250
251 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
252 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
253 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
254 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
255 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
256
257 _, base_client, variant = _split_innertube_client(client)
258 ytcfg['priority'] = 10 * priority(base_client)
259
260 if not variant:
261 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
262 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
263 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
264 embedscreen['priority'] -= 3
265 elif variant == 'embedded':
266 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
267 ytcfg['priority'] -= 2
268 else:
269 ytcfg['priority'] -= 3
270
271
272build_innertube_clients()
273
274
275class YoutubeBaseInfoExtractor(InfoExtractor):
276 """Provide base functions for Youtube extractors"""
277
278 _RESERVED_NAMES = (
279 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
280 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
281 r'browse|oembed|get_video_info|iframe_api|s/player|'
282 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
283
284 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
285
286 # _NETRC_MACHINE = 'youtube'
287
288 # If True it will raise an error if no login info is provided
289 _LOGIN_REQUIRED = False
290
291 _INVIDIOUS_SITES = (
292 # invidious-redirect websites
293 r'(?:www\.)?redirect\.invidious\.io',
294 r'(?:(?:www|dev)\.)?invidio\.us',
295 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
296 r'(?:www\.)?invidious\.pussthecat\.org',
297 r'(?:www\.)?invidious\.zee\.li',
298 r'(?:www\.)?invidious\.ethibox\.fr',
299 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
300 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
301 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
302 # youtube-dl invidious instances list
303 r'(?:(?:www|no)\.)?invidiou\.sh',
304 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
305 r'(?:www\.)?invidious\.kabi\.tk',
306 r'(?:www\.)?invidious\.mastodon\.host',
307 r'(?:www\.)?invidious\.zapashcanon\.fr',
308 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
309 r'(?:www\.)?invidious\.tinfoil-hat\.net',
310 r'(?:www\.)?invidious\.himiko\.cloud',
311 r'(?:www\.)?invidious\.reallyancient\.tech',
312 r'(?:www\.)?invidious\.tube',
313 r'(?:www\.)?invidiou\.site',
314 r'(?:www\.)?invidious\.site',
315 r'(?:www\.)?invidious\.xyz',
316 r'(?:www\.)?invidious\.nixnet\.xyz',
317 r'(?:www\.)?invidious\.048596\.xyz',
318 r'(?:www\.)?invidious\.drycat\.fr',
319 r'(?:www\.)?inv\.skyn3t\.in',
320 r'(?:www\.)?tube\.poal\.co',
321 r'(?:www\.)?tube\.connect\.cafe',
322 r'(?:www\.)?vid\.wxzm\.sx',
323 r'(?:www\.)?vid\.mint\.lgbt',
324 r'(?:www\.)?vid\.puffyan\.us',
325 r'(?:www\.)?yewtu\.be',
326 r'(?:www\.)?yt\.elukerio\.org',
327 r'(?:www\.)?yt\.lelux\.fi',
328 r'(?:www\.)?invidious\.ggc-project\.de',
329 r'(?:www\.)?yt\.maisputain\.ovh',
330 r'(?:www\.)?ytprivate\.com',
331 r'(?:www\.)?invidious\.13ad\.de',
332 r'(?:www\.)?invidious\.toot\.koeln',
333 r'(?:www\.)?invidious\.fdn\.fr',
334 r'(?:www\.)?watch\.nettohikari\.com',
335 r'(?:www\.)?invidious\.namazso\.eu',
336 r'(?:www\.)?invidious\.silkky\.cloud',
337 r'(?:www\.)?invidious\.exonip\.de',
338 r'(?:www\.)?invidious\.riverside\.rocks',
339 r'(?:www\.)?invidious\.blamefran\.net',
340 r'(?:www\.)?invidious\.moomoo\.de',
341 r'(?:www\.)?ytb\.trom\.tf',
342 r'(?:www\.)?yt\.cyberhost\.uk',
343 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
344 r'(?:www\.)?qklhadlycap4cnod\.onion',
345 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
346 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
347 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
348 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
349 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
350 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
351 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
352 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
353 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
354 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
355 )
356
357 def _initialize_consent(self):
358 cookies = self._get_cookies('https://www.youtube.com/')
359 if cookies.get('__Secure-3PSID'):
360 return
361 consent_id = None
362 consent = cookies.get('CONSENT')
363 if consent:
364 if 'YES' in consent.value:
365 return
366 consent_id = self._search_regex(
367 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
368 if not consent_id:
369 consent_id = random.randint(100, 999)
370 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
371
372 def _initialize_pref(self):
373 cookies = self._get_cookies('https://www.youtube.com/')
374 pref_cookie = cookies.get('PREF')
375 pref = {}
376 if pref_cookie:
377 try:
378 pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
379 except ValueError:
380 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
381 pref.update({'hl': 'en', 'tz': 'UTC'})
382 self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
383
384 def _real_initialize(self):
385 self._initialize_pref()
386 self._initialize_consent()
387 if (self._LOGIN_REQUIRED
388 and self.get_param('cookiefile') is None
389 and self.get_param('cookiesfrombrowser') is None):
390 self.raise_login_required('Login details are needed to download this content', method='cookies')
391
392 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
393 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
394 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
395
396 def _get_default_ytcfg(self, client='web'):
397 return copy.deepcopy(INNERTUBE_CLIENTS[client])
398
399 def _get_innertube_host(self, client='web'):
400 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
401
402 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
403 # try_get but with fallback to default ytcfg client values when present
404 _func = lambda y: try_get(y, getter, expected_type)
405 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
406
407 def _extract_client_name(self, ytcfg, default_client='web'):
408 return self._ytcfg_get_safe(
409 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
410 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
411
412 def _extract_client_version(self, ytcfg, default_client='web'):
413 return self._ytcfg_get_safe(
414 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
415 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
416
417 def _extract_api_key(self, ytcfg=None, default_client='web'):
418 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
419
420 def _extract_context(self, ytcfg=None, default_client='web'):
421 context = get_first(
422 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
423 # Enforce language and tz for extraction
424 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
425 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
426 return context
427
428 _SAPISID = None
429
430 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
431 time_now = round(time.time())
432 if self._SAPISID is None:
433 yt_cookies = self._get_cookies('https://www.youtube.com')
434 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
435 # See: https://github.com/yt-dlp/yt-dlp/issues/393
436 sapisid_cookie = dict_get(
437 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
438 if sapisid_cookie and sapisid_cookie.value:
439 self._SAPISID = sapisid_cookie.value
440 self.write_debug('Extracted SAPISID cookie')
441 # SAPISID cookie is required if not already present
442 if not yt_cookies.get('SAPISID'):
443 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
444 self._set_cookie(
445 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
446 else:
447 self._SAPISID = False
448 if not self._SAPISID:
449 return None
450 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
451 sapisidhash = hashlib.sha1(
452 f'{time_now} {self._SAPISID} {origin}'.encode('utf-8')).hexdigest()
453 return f'SAPISIDHASH {time_now}_{sapisidhash}'
454
455 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
456 note='Downloading API JSON', errnote='Unable to download API page',
457 context=None, api_key=None, api_hostname=None, default_client='web'):
458
459 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
460 data.update(query)
461 real_headers = self.generate_api_headers(default_client=default_client)
462 real_headers.update({'content-type': 'application/json'})
463 if headers:
464 real_headers.update(headers)
465 return self._download_json(
466 'https://%s/youtubei/v1/%s' % (api_hostname or self._get_innertube_host(default_client), ep),
467 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
468 data=json.dumps(data).encode('utf8'), headers=real_headers,
469 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
470
471 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
472 data = self._search_regex(
473 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
474 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
475 if data:
476 return self._parse_json(data, item_id, fatal=fatal)
477
478 @staticmethod
479 def _extract_session_index(*data):
480 """
481 Index of current account in account list.
482 See: https://github.com/yt-dlp/yt-dlp/pull/519
483 """
484 for ytcfg in data:
485 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
486 if session_index is not None:
487 return session_index
488
489 # Deprecated?
490 def _extract_identity_token(self, ytcfg=None, webpage=None):
491 if ytcfg:
492 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
493 if token:
494 return token
495 if webpage:
496 return self._search_regex(
497 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
498 'identity token', default=None, fatal=False)
499
500 @staticmethod
501 def _extract_account_syncid(*args):
502 """
503 Extract syncId required to download private playlists of secondary channels
504 @params response and/or ytcfg
505 """
506 for data in args:
507 # ytcfg includes channel_syncid if on secondary channel
508 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
509 if delegated_sid:
510 return delegated_sid
511 sync_ids = (try_get(
512 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
513 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
514 if len(sync_ids) >= 2 and sync_ids[1]:
515 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
516 # and just "user_syncid||" for primary channel. We only want the channel_syncid
517 return sync_ids[0]
518
519 @staticmethod
520 def _extract_visitor_data(*args):
521 """
522 Extracts visitorData from an API response or ytcfg
523 Appears to be used to track session state
524 """
525 return get_first(
526 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
527 expected_type=str)
528
529 @property
530 def is_authenticated(self):
531 return bool(self._generate_sapisidhash_header())
532
533 def extract_ytcfg(self, video_id, webpage):
534 if not webpage:
535 return {}
536 return self._parse_json(
537 self._search_regex(
538 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
539 default='{}'), video_id, fatal=False) or {}
540
541 def generate_api_headers(
542 self, *, ytcfg=None, account_syncid=None, session_index=None,
543 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
544
545 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
546 headers = {
547 'X-YouTube-Client-Name': compat_str(
548 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
549 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
550 'Origin': origin,
551 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
552 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
553 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
554 }
555 if session_index is None:
556 session_index = self._extract_session_index(ytcfg)
557 if account_syncid or session_index is not None:
558 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
559
560 auth = self._generate_sapisidhash_header(origin)
561 if auth is not None:
562 headers['Authorization'] = auth
563 headers['X-Origin'] = origin
564 return {h: v for h, v in headers.items() if v is not None}
565
566 @staticmethod
567 def _build_api_continuation_query(continuation, ctp=None):
568 query = {
569 'continuation': continuation
570 }
571 # TODO: Inconsistency with clickTrackingParams.
572 # Currently we have a fixed ctp contained within context (from ytcfg)
573 # and a ctp in root query for continuation.
574 if ctp:
575 query['clickTracking'] = {'clickTrackingParams': ctp}
576 return query
577
578 @classmethod
579 def _extract_next_continuation_data(cls, renderer):
580 next_continuation = try_get(
581 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
582 lambda x: x['continuation']['reloadContinuationData']), dict)
583 if not next_continuation:
584 return
585 continuation = next_continuation.get('continuation')
586 if not continuation:
587 return
588 ctp = next_continuation.get('clickTrackingParams')
589 return cls._build_api_continuation_query(continuation, ctp)
590
591 @classmethod
592 def _extract_continuation_ep_data(cls, continuation_ep: dict):
593 if isinstance(continuation_ep, dict):
594 continuation = try_get(
595 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
596 if not continuation:
597 return
598 ctp = continuation_ep.get('clickTrackingParams')
599 return cls._build_api_continuation_query(continuation, ctp)
600
601 @classmethod
602 def _extract_continuation(cls, renderer):
603 next_continuation = cls._extract_next_continuation_data(renderer)
604 if next_continuation:
605 return next_continuation
606
607 contents = []
608 for key in ('contents', 'items'):
609 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
610
611 for content in contents:
612 if not isinstance(content, dict):
613 continue
614 continuation_ep = try_get(
615 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
616 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
617 dict)
618 continuation = cls._extract_continuation_ep_data(continuation_ep)
619 if continuation:
620 return continuation
621
622 @classmethod
623 def _extract_alerts(cls, data):
624 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
625 if not isinstance(alert_dict, dict):
626 continue
627 for alert in alert_dict.values():
628 alert_type = alert.get('type')
629 if not alert_type:
630 continue
631 message = cls._get_text(alert, 'text')
632 if message:
633 yield alert_type, message
634
635 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
636 errors = []
637 warnings = []
638 for alert_type, alert_message in alerts:
639 if alert_type.lower() == 'error' and fatal:
640 errors.append([alert_type, alert_message])
641 else:
642 warnings.append([alert_type, alert_message])
643
644 for alert_type, alert_message in (warnings + errors[:-1]):
645 self.report_warning('YouTube said: %s - %s' % (alert_type, alert_message), only_once=only_once)
646 if errors:
647 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
648
649 def _extract_and_report_alerts(self, data, *args, **kwargs):
650 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
651
652 def _extract_badges(self, renderer: dict):
653 badges = set()
654 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
655 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
656 if label:
657 badges.add(label.lower())
658 return badges
659
660 @staticmethod
661 def _get_text(data, *path_list, max_runs=None):
662 for path in path_list or [None]:
663 if path is None:
664 obj = [data]
665 else:
666 obj = traverse_obj(data, path, default=[])
667 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
668 obj = [obj]
669 for item in obj:
670 text = try_get(item, lambda x: x['simpleText'], compat_str)
671 if text:
672 return text
673 runs = try_get(item, lambda x: x['runs'], list) or []
674 if not runs and isinstance(item, list):
675 runs = item
676
677 runs = runs[:min(len(runs), max_runs or len(runs))]
678 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
679 if text:
680 return text
681
682 def _get_count(self, data, *path_list):
683 count_text = self._get_text(data, *path_list) or ''
684 count = parse_count(count_text)
685 if count is None:
686 count = str_to_int(
687 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
688 return count
689
690 @staticmethod
691 def _extract_thumbnails(data, *path_list):
692 """
693 Extract thumbnails from thumbnails dict
694 @param path_list: path list to level that contains 'thumbnails' key
695 """
696 thumbnails = []
697 for path in path_list or [()]:
698 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
699 thumbnail_url = url_or_none(thumbnail.get('url'))
700 if not thumbnail_url:
701 continue
702 # Sometimes youtube gives a wrong thumbnail URL. See:
703 # https://github.com/yt-dlp/yt-dlp/issues/233
704 # https://github.com/ytdl-org/youtube-dl/issues/28023
705 if 'maxresdefault' in thumbnail_url:
706 thumbnail_url = thumbnail_url.split('?')[0]
707 thumbnails.append({
708 'url': thumbnail_url,
709 'height': int_or_none(thumbnail.get('height')),
710 'width': int_or_none(thumbnail.get('width')),
711 })
712 return thumbnails
713
714 @staticmethod
715 def extract_relative_time(relative_time_text):
716 """
717 Extracts a relative time from string and converts to dt object
718 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
719 """
720 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)
721 if mobj:
722 start = mobj.group('start')
723 if start:
724 return datetime_from_str(start)
725 try:
726 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
727 except ValueError:
728 return None
729
730 def _extract_time_text(self, renderer, *path_list):
731 text = self._get_text(renderer, *path_list) or ''
732 dt = self.extract_relative_time(text)
733 timestamp = None
734 if isinstance(dt, datetime.datetime):
735 timestamp = calendar.timegm(dt.timetuple())
736
737 if timestamp is None:
738 timestamp = (
739 unified_timestamp(text) or unified_timestamp(
740 self._search_regex(
741 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
742 text.lower(), 'time text', default=None)))
743
744 if text and timestamp is None:
745 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
746 return timestamp, text
747
748 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
749 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
750 default_client='web'):
751 response = None
752 last_error = None
753 count = -1
754 retries = self.get_param('extractor_retries', 3)
755 if check_get_keys is None:
756 check_get_keys = []
757 while count < retries:
758 count += 1
759 if last_error:
760 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
761 try:
762 response = self._call_api(
763 ep=ep, fatal=True, headers=headers,
764 video_id=item_id, query=query,
765 context=self._extract_context(ytcfg, default_client),
766 api_key=self._extract_api_key(ytcfg, default_client),
767 api_hostname=api_hostname, default_client=default_client,
768 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
769 except ExtractorError as e:
770 if isinstance(e.cause, network_exceptions):
771 if isinstance(e.cause, compat_HTTPError):
772 first_bytes = e.cause.read(512)
773 if not is_html(first_bytes):
774 yt_error = try_get(
775 self._parse_json(
776 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
777 lambda x: x['error']['message'], compat_str)
778 if yt_error:
779 self._report_alerts([('ERROR', yt_error)], fatal=False)
780 # Downloading page may result in intermittent 5xx HTTP error
781 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
782 # We also want to catch all other network exceptions since errors in later pages can be troublesome
783 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
784 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
785 last_error = error_to_compat_str(e.cause or e.msg)
786 if count < retries:
787 continue
788 if fatal:
789 raise
790 else:
791 self.report_warning(error_to_compat_str(e))
792 return
793
794 else:
795 try:
796 self._extract_and_report_alerts(response, only_once=True)
797 except ExtractorError as e:
798 # YouTube servers may return errors we want to retry on in a 200 OK response
799 # See: https://github.com/yt-dlp/yt-dlp/issues/839
800 if 'unknown error' in e.msg.lower():
801 last_error = e.msg
802 continue
803 if fatal:
804 raise
805 self.report_warning(error_to_compat_str(e))
806 return
807 if not check_get_keys or dict_get(response, check_get_keys):
808 break
809 # Youtube sometimes sends incomplete data
810 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
811 last_error = 'Incomplete data received'
812 if count >= retries:
813 if fatal:
814 raise ExtractorError(last_error)
815 else:
816 self.report_warning(last_error)
817 return
818 return response
819
820 @staticmethod
821 def is_music_url(url):
822 return re.match(r'https?://music\.youtube\.com/', url) is not None
823
824 def _extract_video(self, renderer):
825 video_id = renderer.get('videoId')
826 title = self._get_text(renderer, 'title')
827 description = self._get_text(renderer, 'descriptionSnippet')
828 duration = parse_duration(self._get_text(
829 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
830 if duration is None:
831 duration = parse_duration(self._search_regex(
832 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
833 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
834 video_id, default=None, group='duration'))
835
836 view_count = self._get_count(renderer, 'viewCountText')
837
838 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
839 channel_id = traverse_obj(
840 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'), expected_type=str, get_all=False)
841 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
842 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
843 overlay_style = traverse_obj(
844 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'), get_all=False, expected_type=str)
845 badges = self._extract_badges(renderer)
846 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
847 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
848 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'), expected_type=str))
849 url = f'https://www.youtube.com/watch?v={video_id}'
850 if overlay_style == 'SHORTS' or (navigation_url and '/shorts/' in navigation_url):
851 url = f'https://www.youtube.com/shorts/{video_id}'
852
853 return {
854 '_type': 'url',
855 'ie_key': YoutubeIE.ie_key(),
856 'id': video_id,
857 'url': url,
858 'title': title,
859 'description': description,
860 'duration': duration,
861 'view_count': view_count,
862 'uploader': uploader,
863 'channel_id': channel_id,
864 'thumbnails': thumbnails,
865 'upload_date': strftime_or_none(timestamp, '%Y%m%d') if self._configuration_arg('approximate_date', ie_key='youtubetab') else None,
866 'live_status': ('is_upcoming' if scheduled_timestamp is not None
867 else 'was_live' if 'streamed' in time_text.lower()
868 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
869 else None),
870 'release_timestamp': scheduled_timestamp,
871 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
872 }
873
874
875class YoutubeIE(YoutubeBaseInfoExtractor):
876 IE_DESC = 'YouTube'
877 _VALID_URL = r"""(?x)^
878 (
879 (?:https?://|//) # http(s):// or protocol-independent URL
880 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
881 (?:www\.)?deturl\.com/www\.youtube\.com|
882 (?:www\.)?pwnyoutube\.com|
883 (?:www\.)?hooktube\.com|
884 (?:www\.)?yourepeat\.com|
885 tube\.majestyc\.net|
886 %(invidious)s|
887 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
888 (?:.*?\#/)? # handle anchor (#/) redirect urls
889 (?: # the various things that can precede the ID:
890 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
891 |(?: # or the v= param in all its forms
892 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
893 (?:\?|\#!?) # the params delimiter ? or # or #!
894 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
895 v=
896 )
897 ))
898 |(?:
899 youtu\.be| # just youtu.be/xxxx
900 vid\.plus| # or vid.plus/xxxx
901 zwearz\.com/watch| # or zwearz.com/watch/xxxx
902 %(invidious)s
903 )/
904 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
905 )
906 )? # all until now is optional -> you can pass the naked ID
907 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
908 (?(1).+)? # if we found the ID, everything can follow
909 (?:\#|$)""" % {
910 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
911 }
912 _PLAYER_INFO_RE = (
913 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
914 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$',
915 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
916 )
917 _formats = {
918 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
919 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
920 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
921 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
922 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
923 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
924 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
925 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
926 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
927 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
928 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
929 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
930 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
931 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
932 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
933 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
934 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
935 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
936
937
938 # 3D videos
939 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
940 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
941 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
942 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
943 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
944 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
945 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
946
947 # Apple HTTP Live Streaming
948 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
949 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
950 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
951 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
952 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
953 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
954 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
955 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
956
957 # DASH mp4 video
958 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
959 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
960 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
961 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
962 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
963 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
964 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
965 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
966 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
967 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
968 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
969 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
970
971 # Dash mp4 audio
972 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
973 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
974 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
975 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
976 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
977 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
978 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
979
980 # Dash webm
981 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
982 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
983 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
984 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
985 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
986 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
987 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
988 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
989 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
990 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
991 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
992 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
993 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
994 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
995 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
996 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
997 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
998 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
999 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1000 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1001 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1002 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1003
1004 # Dash webm audio
1005 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1006 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1007
1008 # Dash webm audio with opus inside
1009 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1010 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1011 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1012
1013 # RTMP (unnamed)
1014 '_rtmp': {'protocol': 'rtmp'},
1015
1016 # av01 video only formats sometimes served with "unknown" codecs
1017 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1018 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1019 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1020 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1021 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1022 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1023 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1024 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1025 }
1026 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1027
1028 _GEO_BYPASS = False
1029
1030 IE_NAME = 'youtube'
1031 _TESTS = [
1032 {
1033 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1034 'info_dict': {
1035 'id': 'BaW_jenozKc',
1036 'ext': 'mp4',
1037 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1038 'uploader': 'Philipp Hagemeister',
1039 'uploader_id': 'phihag',
1040 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1041 'channel': 'Philipp Hagemeister',
1042 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1043 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1044 'upload_date': '20121002',
1045 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1046 'categories': ['Science & Technology'],
1047 'tags': ['youtube-dl'],
1048 'duration': 10,
1049 'view_count': int,
1050 'like_count': int,
1051 'availability': 'public',
1052 'playable_in_embed': True,
1053 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1054 'live_status': 'not_live',
1055 'age_limit': 0,
1056 'start_time': 1,
1057 'end_time': 9,
1058 'channel_follower_count': int
1059 }
1060 },
1061 {
1062 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1063 'note': 'Embed-only video (#1746)',
1064 'info_dict': {
1065 'id': 'yZIXLfi8CZQ',
1066 'ext': 'mp4',
1067 'upload_date': '20120608',
1068 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1069 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1070 'uploader': 'SET India',
1071 'uploader_id': 'setindia',
1072 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1073 'age_limit': 18,
1074 },
1075 'skip': 'Private video',
1076 },
1077 {
1078 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1079 'note': 'Use the first video ID in the URL',
1080 'info_dict': {
1081 'id': 'BaW_jenozKc',
1082 'ext': 'mp4',
1083 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1084 'uploader': 'Philipp Hagemeister',
1085 'uploader_id': 'phihag',
1086 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1087 'channel': 'Philipp Hagemeister',
1088 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1089 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1090 'upload_date': '20121002',
1091 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1092 'categories': ['Science & Technology'],
1093 'tags': ['youtube-dl'],
1094 'duration': 10,
1095 'view_count': int,
1096 'like_count': int,
1097 'availability': 'public',
1098 'playable_in_embed': True,
1099 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1100 'live_status': 'not_live',
1101 'age_limit': 0,
1102 'channel_follower_count': int
1103 },
1104 'params': {
1105 'skip_download': True,
1106 },
1107 },
1108 {
1109 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1110 'note': '256k DASH audio (format 141) via DASH manifest',
1111 'info_dict': {
1112 'id': 'a9LDPn-MO4I',
1113 'ext': 'm4a',
1114 'upload_date': '20121002',
1115 'uploader_id': '8KVIDEO',
1116 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1117 'description': '',
1118 'uploader': '8KVIDEO',
1119 'title': 'UHDTV TEST 8K VIDEO.mp4'
1120 },
1121 'params': {
1122 'youtube_include_dash_manifest': True,
1123 'format': '141',
1124 },
1125 'skip': 'format 141 not served anymore',
1126 },
1127 # DASH manifest with encrypted signature
1128 {
1129 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1130 'info_dict': {
1131 'id': 'IB3lcPjvWLA',
1132 'ext': 'm4a',
1133 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1134 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1135 'duration': 244,
1136 'uploader': 'AfrojackVEVO',
1137 'uploader_id': 'AfrojackVEVO',
1138 'upload_date': '20131011',
1139 'abr': 129.495,
1140 'like_count': int,
1141 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1142 'playable_in_embed': True,
1143 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1144 'view_count': int,
1145 'track': 'The Spark',
1146 'live_status': 'not_live',
1147 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1148 'channel': 'Afrojack',
1149 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1150 'tags': 'count:19',
1151 'availability': 'public',
1152 'categories': ['Music'],
1153 'age_limit': 0,
1154 'alt_title': 'The Spark',
1155 'channel_follower_count': int
1156 },
1157 'params': {
1158 'youtube_include_dash_manifest': True,
1159 'format': '141/bestaudio[ext=m4a]',
1160 },
1161 },
1162 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1163 {
1164 'note': 'Embed allowed age-gate video',
1165 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1166 'info_dict': {
1167 'id': 'HtVdAasjOgU',
1168 'ext': 'mp4',
1169 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1170 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1171 'duration': 142,
1172 'uploader': 'The Witcher',
1173 'uploader_id': 'WitcherGame',
1174 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1175 'upload_date': '20140605',
1176 'age_limit': 18,
1177 'categories': ['Gaming'],
1178 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1179 'availability': 'needs_auth',
1180 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1181 'like_count': int,
1182 'channel': 'The Witcher',
1183 'live_status': 'not_live',
1184 'tags': 'count:17',
1185 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1186 'playable_in_embed': True,
1187 'view_count': int,
1188 'channel_follower_count': int
1189 },
1190 },
1191 {
1192 'note': 'Age-gate video with embed allowed in public site',
1193 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1194 'info_dict': {
1195 'id': 'HsUATh_Nc2U',
1196 'ext': 'mp4',
1197 'title': 'Godzilla 2 (Official Video)',
1198 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1199 'upload_date': '20200408',
1200 'uploader_id': 'FlyingKitty900',
1201 'uploader': 'FlyingKitty',
1202 'age_limit': 18,
1203 'availability': 'needs_auth',
1204 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1205 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1206 'channel': 'FlyingKitty',
1207 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1208 'view_count': int,
1209 'categories': ['Entertainment'],
1210 'live_status': 'not_live',
1211 'tags': ['Flyingkitty', 'godzilla 2'],
1212 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1213 'like_count': int,
1214 'duration': 177,
1215 'playable_in_embed': True,
1216 'channel_follower_count': int
1217 },
1218 },
1219 {
1220 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1221 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1222 'info_dict': {
1223 'id': 'Tq92D6wQ1mg',
1224 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1225 'ext': 'mp4',
1226 'upload_date': '20191228',
1227 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1228 'uploader': 'Projekt Melody',
1229 'description': 'md5:17eccca93a786d51bc67646756894066',
1230 'age_limit': 18,
1231 'like_count': int,
1232 'availability': 'needs_auth',
1233 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1234 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1235 'view_count': int,
1236 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1237 'channel': 'Projekt Melody',
1238 'live_status': 'not_live',
1239 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1240 'playable_in_embed': True,
1241 'categories': ['Entertainment'],
1242 'duration': 106,
1243 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1244 'channel_follower_count': int
1245 },
1246 },
1247 {
1248 'note': 'Non-Agegated non-embeddable video',
1249 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1250 'info_dict': {
1251 'id': 'MeJVWBSsPAY',
1252 'ext': 'mp4',
1253 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1254 'uploader': 'Herr Lurik',
1255 'uploader_id': 'st3in234',
1256 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1257 'upload_date': '20130730',
1258 'track': 'Such mich find mich',
1259 'age_limit': 0,
1260 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1261 'like_count': int,
1262 'playable_in_embed': False,
1263 'creator': 'OOMPH!',
1264 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1265 'view_count': int,
1266 'alt_title': 'Such mich find mich',
1267 'duration': 210,
1268 'channel': 'Herr Lurik',
1269 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1270 'categories': ['Music'],
1271 'availability': 'public',
1272 'uploader_url': 'http://www.youtube.com/user/st3in234',
1273 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1274 'live_status': 'not_live',
1275 'artist': 'OOMPH!',
1276 'channel_follower_count': int
1277 },
1278 },
1279 {
1280 'note': 'Non-bypassable age-gated video',
1281 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1282 'only_matching': True,
1283 },
1284 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1285 # YouTube Red ad is not captured for creator
1286 {
1287 'url': '__2ABJjxzNo',
1288 'info_dict': {
1289 'id': '__2ABJjxzNo',
1290 'ext': 'mp4',
1291 'duration': 266,
1292 'upload_date': '20100430',
1293 'uploader_id': 'deadmau5',
1294 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1295 'creator': 'deadmau5',
1296 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1297 'uploader': 'deadmau5',
1298 'title': 'Deadmau5 - Some Chords (HD)',
1299 'alt_title': 'Some Chords',
1300 'availability': 'public',
1301 'tags': 'count:14',
1302 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1303 'view_count': int,
1304 'live_status': 'not_live',
1305 'channel': 'deadmau5',
1306 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1307 'like_count': int,
1308 'track': 'Some Chords',
1309 'artist': 'deadmau5',
1310 'playable_in_embed': True,
1311 'age_limit': 0,
1312 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1313 'categories': ['Music'],
1314 'album': 'Some Chords',
1315 'channel_follower_count': int
1316 },
1317 'expected_warnings': [
1318 'DASH manifest missing',
1319 ]
1320 },
1321 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1322 {
1323 'url': 'lqQg6PlCWgI',
1324 'info_dict': {
1325 'id': 'lqQg6PlCWgI',
1326 'ext': 'mp4',
1327 'duration': 6085,
1328 'upload_date': '20150827',
1329 'uploader_id': 'olympic',
1330 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1331 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1332 'uploader': 'Olympics',
1333 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1334 'like_count': int,
1335 'release_timestamp': 1343767800,
1336 'playable_in_embed': True,
1337 'categories': ['Sports'],
1338 'release_date': '20120731',
1339 'channel': 'Olympics',
1340 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1341 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1342 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1343 'age_limit': 0,
1344 'availability': 'public',
1345 'live_status': 'was_live',
1346 'view_count': int,
1347 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1348 'channel_follower_count': int
1349 },
1350 'params': {
1351 'skip_download': 'requires avconv',
1352 }
1353 },
1354 # Non-square pixels
1355 {
1356 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1357 'info_dict': {
1358 'id': '_b-2C3KPAM0',
1359 'ext': 'mp4',
1360 'stretched_ratio': 16 / 9.,
1361 'duration': 85,
1362 'upload_date': '20110310',
1363 'uploader_id': 'AllenMeow',
1364 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1365 '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',
1366 'uploader': '孫ᄋᄅ',
1367 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1368 'playable_in_embed': True,
1369 'channel': '孫ᄋᄅ',
1370 'age_limit': 0,
1371 'tags': 'count:11',
1372 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1373 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1374 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1375 'view_count': int,
1376 'categories': ['People & Blogs'],
1377 'like_count': int,
1378 'live_status': 'not_live',
1379 'availability': 'unlisted',
1380 'channel_follower_count': int
1381 },
1382 },
1383 # url_encoded_fmt_stream_map is empty string
1384 {
1385 'url': 'qEJwOuvDf7I',
1386 'info_dict': {
1387 'id': 'qEJwOuvDf7I',
1388 'ext': 'webm',
1389 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1390 'description': '',
1391 'upload_date': '20150404',
1392 'uploader_id': 'spbelect',
1393 'uploader': 'Наблюдатели Петербурга',
1394 },
1395 'params': {
1396 'skip_download': 'requires avconv',
1397 },
1398 'skip': 'This live event has ended.',
1399 },
1400 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1401 {
1402 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1403 'info_dict': {
1404 'id': 'FIl7x6_3R5Y',
1405 'ext': 'webm',
1406 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1407 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1408 'duration': 220,
1409 'upload_date': '20150625',
1410 'uploader_id': 'dorappi2000',
1411 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1412 'uploader': 'dorappi2000',
1413 'formats': 'mincount:31',
1414 },
1415 'skip': 'not actual anymore',
1416 },
1417 # DASH manifest with segment_list
1418 {
1419 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1420 'md5': '8ce563a1d667b599d21064e982ab9e31',
1421 'info_dict': {
1422 'id': 'CsmdDsKjzN8',
1423 'ext': 'mp4',
1424 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1425 'uploader': 'Airtek',
1426 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1427 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1428 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1429 },
1430 'params': {
1431 'youtube_include_dash_manifest': True,
1432 'format': '135', # bestvideo
1433 },
1434 'skip': 'This live event has ended.',
1435 },
1436 {
1437 # Multifeed videos (multiple cameras), URL is for Main Camera
1438 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1439 'info_dict': {
1440 'id': 'jvGDaLqkpTg',
1441 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1442 'description': 'md5:e03b909557865076822aa169218d6a5d',
1443 },
1444 'playlist': [{
1445 'info_dict': {
1446 'id': 'jvGDaLqkpTg',
1447 'ext': 'mp4',
1448 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1449 'description': 'md5:e03b909557865076822aa169218d6a5d',
1450 'duration': 10643,
1451 'upload_date': '20161111',
1452 'uploader': 'Team PGP',
1453 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1454 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1455 },
1456 }, {
1457 'info_dict': {
1458 'id': '3AKt1R1aDnw',
1459 'ext': 'mp4',
1460 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1461 'description': 'md5:e03b909557865076822aa169218d6a5d',
1462 'duration': 10991,
1463 'upload_date': '20161111',
1464 'uploader': 'Team PGP',
1465 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1466 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1467 },
1468 }, {
1469 'info_dict': {
1470 'id': 'RtAMM00gpVc',
1471 'ext': 'mp4',
1472 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1473 'description': 'md5:e03b909557865076822aa169218d6a5d',
1474 'duration': 10995,
1475 'upload_date': '20161111',
1476 'uploader': 'Team PGP',
1477 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1478 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1479 },
1480 }, {
1481 'info_dict': {
1482 'id': '6N2fdlP3C5U',
1483 'ext': 'mp4',
1484 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1485 'description': 'md5:e03b909557865076822aa169218d6a5d',
1486 'duration': 10990,
1487 'upload_date': '20161111',
1488 'uploader': 'Team PGP',
1489 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1490 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1491 },
1492 }],
1493 'params': {
1494 'skip_download': True,
1495 },
1496 'skip': 'Not multifeed anymore',
1497 },
1498 {
1499 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1500 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1501 'info_dict': {
1502 'id': 'gVfLd0zydlo',
1503 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1504 },
1505 'playlist_count': 2,
1506 'skip': 'Not multifeed anymore',
1507 },
1508 {
1509 'url': 'https://vid.plus/FlRa-iH7PGw',
1510 'only_matching': True,
1511 },
1512 {
1513 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1514 'only_matching': True,
1515 },
1516 {
1517 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1518 # Also tests cut-off URL expansion in video description (see
1519 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1520 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1521 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1522 'info_dict': {
1523 'id': 'lsguqyKfVQg',
1524 'ext': 'mp4',
1525 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1526 'alt_title': 'Dark Walk',
1527 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1528 'duration': 133,
1529 'upload_date': '20151119',
1530 'uploader_id': 'IronSoulElf',
1531 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1532 'uploader': 'IronSoulElf',
1533 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1534 'track': 'Dark Walk',
1535 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1536 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1537 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1538 'categories': ['Film & Animation'],
1539 'view_count': int,
1540 'live_status': 'not_live',
1541 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1542 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1543 'tags': 'count:13',
1544 'availability': 'public',
1545 'channel': 'IronSoulElf',
1546 'playable_in_embed': True,
1547 'like_count': int,
1548 'age_limit': 0,
1549 'channel_follower_count': int
1550 },
1551 'params': {
1552 'skip_download': True,
1553 },
1554 },
1555 {
1556 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1557 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1558 'only_matching': True,
1559 },
1560 {
1561 # Video with yt:stretch=17:0
1562 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1563 'info_dict': {
1564 'id': 'Q39EVAstoRM',
1565 'ext': 'mp4',
1566 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1567 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1568 'upload_date': '20151107',
1569 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1570 'uploader': 'CH GAMER DROID',
1571 },
1572 'params': {
1573 'skip_download': True,
1574 },
1575 'skip': 'This video does not exist.',
1576 },
1577 {
1578 # Video with incomplete 'yt:stretch=16:'
1579 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1580 'only_matching': True,
1581 },
1582 {
1583 # Video licensed under Creative Commons
1584 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1585 'info_dict': {
1586 'id': 'M4gD1WSo5mA',
1587 'ext': 'mp4',
1588 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1589 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1590 'duration': 721,
1591 'upload_date': '20150128',
1592 'uploader_id': 'BerkmanCenter',
1593 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1594 'uploader': 'The Berkman Klein Center for Internet & Society',
1595 'license': 'Creative Commons Attribution license (reuse allowed)',
1596 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1597 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1598 'like_count': int,
1599 'age_limit': 0,
1600 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1601 'channel': 'The Berkman Klein Center for Internet & Society',
1602 'availability': 'public',
1603 'view_count': int,
1604 'categories': ['Education'],
1605 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1606 'live_status': 'not_live',
1607 'playable_in_embed': True,
1608 'channel_follower_count': int
1609 },
1610 'params': {
1611 'skip_download': True,
1612 },
1613 },
1614 {
1615 # Channel-like uploader_url
1616 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1617 'info_dict': {
1618 'id': 'eQcmzGIKrzg',
1619 'ext': 'mp4',
1620 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1621 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1622 'duration': 4060,
1623 'upload_date': '20151120',
1624 'uploader': 'Bernie Sanders',
1625 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1626 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1627 'license': 'Creative Commons Attribution license (reuse allowed)',
1628 'playable_in_embed': True,
1629 'tags': 'count:12',
1630 'like_count': int,
1631 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1632 'age_limit': 0,
1633 'availability': 'public',
1634 'categories': ['News & Politics'],
1635 'channel': 'Bernie Sanders',
1636 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1637 'view_count': int,
1638 'live_status': 'not_live',
1639 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1640 'channel_follower_count': int
1641 },
1642 'params': {
1643 'skip_download': True,
1644 },
1645 },
1646 {
1647 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1648 'only_matching': True,
1649 },
1650 {
1651 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1652 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1653 'only_matching': True,
1654 },
1655 {
1656 # Rental video preview
1657 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1658 'info_dict': {
1659 'id': 'uGpuVWrhIzE',
1660 'ext': 'mp4',
1661 'title': 'Piku - Trailer',
1662 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1663 'upload_date': '20150811',
1664 'uploader': 'FlixMatrix',
1665 'uploader_id': 'FlixMatrixKaravan',
1666 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1667 'license': 'Standard YouTube License',
1668 },
1669 'params': {
1670 'skip_download': True,
1671 },
1672 'skip': 'This video is not available.',
1673 },
1674 {
1675 # YouTube Red video with episode data
1676 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1677 'info_dict': {
1678 'id': 'iqKdEhx-dD4',
1679 'ext': 'mp4',
1680 'title': 'Isolation - Mind Field (Ep 1)',
1681 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1682 'duration': 2085,
1683 'upload_date': '20170118',
1684 'uploader': 'Vsauce',
1685 'uploader_id': 'Vsauce',
1686 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1687 'series': 'Mind Field',
1688 'season_number': 1,
1689 'episode_number': 1,
1690 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1691 'tags': 'count:12',
1692 'view_count': int,
1693 'availability': 'public',
1694 'age_limit': 0,
1695 'channel': 'Vsauce',
1696 'episode': 'Episode 1',
1697 'categories': ['Entertainment'],
1698 'season': 'Season 1',
1699 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1700 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1701 'like_count': int,
1702 'playable_in_embed': True,
1703 'live_status': 'not_live',
1704 'channel_follower_count': int
1705 },
1706 'params': {
1707 'skip_download': True,
1708 },
1709 'expected_warnings': [
1710 'Skipping DASH manifest',
1711 ],
1712 },
1713 {
1714 # The following content has been identified by the YouTube community
1715 # as inappropriate or offensive to some audiences.
1716 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1717 'info_dict': {
1718 'id': '6SJNVb0GnPI',
1719 'ext': 'mp4',
1720 'title': 'Race Differences in Intelligence',
1721 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1722 'duration': 965,
1723 'upload_date': '20140124',
1724 'uploader': 'New Century Foundation',
1725 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1726 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1727 },
1728 'params': {
1729 'skip_download': True,
1730 },
1731 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1732 },
1733 {
1734 # itag 212
1735 'url': '1t24XAntNCY',
1736 'only_matching': True,
1737 },
1738 {
1739 # geo restricted to JP
1740 'url': 'sJL6WA-aGkQ',
1741 'only_matching': True,
1742 },
1743 {
1744 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1745 'only_matching': True,
1746 },
1747 {
1748 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1749 'only_matching': True,
1750 },
1751 {
1752 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1753 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1754 'only_matching': True,
1755 },
1756 {
1757 # DRM protected
1758 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1759 'only_matching': True,
1760 },
1761 {
1762 # Video with unsupported adaptive stream type formats
1763 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1764 'info_dict': {
1765 'id': 'Z4Vy8R84T1U',
1766 'ext': 'mp4',
1767 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1768 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1769 'duration': 433,
1770 'upload_date': '20130923',
1771 'uploader': 'Amelia Putri Harwita',
1772 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1773 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1774 'formats': 'maxcount:10',
1775 },
1776 'params': {
1777 'skip_download': True,
1778 'youtube_include_dash_manifest': False,
1779 },
1780 'skip': 'not actual anymore',
1781 },
1782 {
1783 # Youtube Music Auto-generated description
1784 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1785 'info_dict': {
1786 'id': 'MgNrAu2pzNs',
1787 'ext': 'mp4',
1788 'title': 'Voyeur Girl',
1789 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1790 'upload_date': '20190312',
1791 'uploader': 'Stephen - Topic',
1792 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1793 'artist': 'Stephen',
1794 'track': 'Voyeur Girl',
1795 'album': 'it\'s too much love to know my dear',
1796 'release_date': '20190313',
1797 'release_year': 2019,
1798 'alt_title': 'Voyeur Girl',
1799 'view_count': int,
1800 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1801 'playable_in_embed': True,
1802 'like_count': int,
1803 'categories': ['Music'],
1804 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1805 'channel': 'Stephen',
1806 'availability': 'public',
1807 'creator': 'Stephen',
1808 'duration': 169,
1809 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1810 'age_limit': 0,
1811 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1812 'tags': 'count:11',
1813 'live_status': 'not_live',
1814 'channel_follower_count': int
1815 },
1816 'params': {
1817 'skip_download': True,
1818 },
1819 },
1820 {
1821 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1822 'only_matching': True,
1823 },
1824 {
1825 # invalid -> valid video id redirection
1826 'url': 'DJztXj2GPfl',
1827 'info_dict': {
1828 'id': 'DJztXj2GPfk',
1829 'ext': 'mp4',
1830 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1831 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1832 'upload_date': '20090125',
1833 'uploader': 'Prochorowka',
1834 'uploader_id': 'Prochorowka',
1835 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1836 'artist': 'Panjabi MC',
1837 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1838 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1839 },
1840 'params': {
1841 'skip_download': True,
1842 },
1843 'skip': 'Video unavailable',
1844 },
1845 {
1846 # empty description results in an empty string
1847 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1848 'info_dict': {
1849 'id': 'x41yOUIvK2k',
1850 'ext': 'mp4',
1851 'title': 'IMG 3456',
1852 'description': '',
1853 'upload_date': '20170613',
1854 'uploader_id': 'ElevageOrVert',
1855 'uploader': 'ElevageOrVert',
1856 'view_count': int,
1857 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1858 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1859 'like_count': int,
1860 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1861 'tags': [],
1862 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1863 'availability': 'public',
1864 'age_limit': 0,
1865 'categories': ['Pets & Animals'],
1866 'duration': 7,
1867 'playable_in_embed': True,
1868 'live_status': 'not_live',
1869 'channel': 'ElevageOrVert',
1870 'channel_follower_count': int
1871 },
1872 'params': {
1873 'skip_download': True,
1874 },
1875 },
1876 {
1877 # with '};' inside yt initial data (see [1])
1878 # see [2] for an example with '};' inside ytInitialPlayerResponse
1879 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1880 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1881 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1882 'info_dict': {
1883 'id': 'CHqg6qOn4no',
1884 'ext': 'mp4',
1885 'title': 'Part 77 Sort a list of simple types in c#',
1886 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1887 'upload_date': '20130831',
1888 'uploader_id': 'kudvenkat',
1889 'uploader': 'kudvenkat',
1890 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1891 'like_count': int,
1892 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1893 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1894 'live_status': 'not_live',
1895 'categories': ['Education'],
1896 'availability': 'public',
1897 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1898 'tags': 'count:12',
1899 'playable_in_embed': True,
1900 'age_limit': 0,
1901 'view_count': int,
1902 'duration': 522,
1903 'channel': 'kudvenkat',
1904 'channel_follower_count': int
1905 },
1906 'params': {
1907 'skip_download': True,
1908 },
1909 },
1910 {
1911 # another example of '};' in ytInitialData
1912 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1913 'only_matching': True,
1914 },
1915 {
1916 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1917 'only_matching': True,
1918 },
1919 {
1920 # https://github.com/ytdl-org/youtube-dl/pull/28094
1921 'url': 'OtqTfy26tG0',
1922 'info_dict': {
1923 'id': 'OtqTfy26tG0',
1924 'ext': 'mp4',
1925 'title': 'Burn Out',
1926 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1927 'upload_date': '20141120',
1928 'uploader': 'The Cinematic Orchestra - Topic',
1929 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1930 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1931 'artist': 'The Cinematic Orchestra',
1932 'track': 'Burn Out',
1933 'album': 'Every Day',
1934 'like_count': int,
1935 'live_status': 'not_live',
1936 'alt_title': 'Burn Out',
1937 'duration': 614,
1938 'age_limit': 0,
1939 'view_count': int,
1940 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1941 'creator': 'The Cinematic Orchestra',
1942 'channel': 'The Cinematic Orchestra',
1943 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1944 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1945 'availability': 'public',
1946 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1947 'categories': ['Music'],
1948 'playable_in_embed': True,
1949 'channel_follower_count': int
1950 },
1951 'params': {
1952 'skip_download': True,
1953 },
1954 },
1955 {
1956 # controversial video, only works with bpctr when authenticated with cookies
1957 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1958 'only_matching': True,
1959 },
1960 {
1961 # controversial video, requires bpctr/contentCheckOk
1962 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1963 'info_dict': {
1964 'id': 'SZJvDhaSDnc',
1965 'ext': 'mp4',
1966 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1967 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1968 'uploader': 'CBS Mornings',
1969 'uploader_id': 'CBSThisMorning',
1970 'upload_date': '20140716',
1971 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
1972 'duration': 170,
1973 'categories': ['News & Politics'],
1974 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
1975 'view_count': int,
1976 'channel': 'CBS Mornings',
1977 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
1978 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
1979 'age_limit': 18,
1980 'availability': 'needs_auth',
1981 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
1982 'like_count': int,
1983 'live_status': 'not_live',
1984 'playable_in_embed': True,
1985 'channel_follower_count': int
1986 }
1987 },
1988 {
1989 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
1990 'url': 'cBvYw8_A0vQ',
1991 'info_dict': {
1992 'id': 'cBvYw8_A0vQ',
1993 'ext': 'mp4',
1994 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
1995 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
1996 'upload_date': '20201120',
1997 'uploader': 'Walk around Japan',
1998 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
1999 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2000 'duration': 1456,
2001 'categories': ['Travel & Events'],
2002 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2003 'view_count': int,
2004 'channel': 'Walk around Japan',
2005 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2006 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2007 'age_limit': 0,
2008 'availability': 'public',
2009 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2010 'live_status': 'not_live',
2011 'playable_in_embed': True,
2012 'channel_follower_count': int
2013 },
2014 'params': {
2015 'skip_download': True,
2016 },
2017 }, {
2018 # Has multiple audio streams
2019 'url': 'WaOKSUlf4TM',
2020 'only_matching': True
2021 }, {
2022 # Requires Premium: has format 141 when requested using YTM url
2023 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2024 'only_matching': True
2025 }, {
2026 # multiple subtitles with same lang_code
2027 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2028 'only_matching': True,
2029 }, {
2030 # Force use android client fallback
2031 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2032 'info_dict': {
2033 'id': 'YOelRv7fMxY',
2034 'title': 'DIGGING A SECRET TUNNEL Part 1',
2035 'ext': '3gp',
2036 'upload_date': '20210624',
2037 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2038 'uploader': 'colinfurze',
2039 'uploader_id': 'colinfurze',
2040 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2041 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2042 'duration': 596,
2043 'categories': ['Entertainment'],
2044 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2045 'view_count': int,
2046 'channel': 'colinfurze',
2047 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2048 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2049 'age_limit': 0,
2050 'availability': 'public',
2051 'like_count': int,
2052 'live_status': 'not_live',
2053 'playable_in_embed': True,
2054 'channel_follower_count': int
2055 },
2056 'params': {
2057 'format': '17', # 3gp format available on android
2058 'extractor_args': {'youtube': {'player_client': ['android']}},
2059 },
2060 },
2061 {
2062 # Skip download of additional client configs (remix client config in this case)
2063 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2064 'only_matching': True,
2065 'params': {
2066 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2067 },
2068 }, {
2069 # shorts
2070 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2071 'only_matching': True,
2072 }, {
2073 'note': 'Storyboards',
2074 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2075 'info_dict': {
2076 'id': '5KLPxDtMqe8',
2077 'ext': 'mhtml',
2078 'format_id': 'sb0',
2079 'title': 'Your Brain is Plastic',
2080 'uploader_id': 'scishow',
2081 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2082 'upload_date': '20140324',
2083 'uploader': 'SciShow',
2084 'like_count': int,
2085 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2086 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2087 'view_count': int,
2088 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2089 'playable_in_embed': True,
2090 'tags': 'count:12',
2091 'uploader_url': 'http://www.youtube.com/user/scishow',
2092 'availability': 'public',
2093 'channel': 'SciShow',
2094 'live_status': 'not_live',
2095 'duration': 248,
2096 'categories': ['Education'],
2097 'age_limit': 0,
2098 'channel_follower_count': int
2099 }, 'params': {'format': 'mhtml', 'skip_download': True}
2100 }, {
2101 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2102 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2103 'info_dict': {
2104 'id': '2NUZ8W2llS4',
2105 'ext': 'mp4',
2106 'title': 'The NP that test your phone performance 🙂',
2107 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2108 'uploader': 'Leon Nguyen',
2109 'uploader_id': 'VNSXIII',
2110 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2111 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2112 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2113 'duration': 21,
2114 'view_count': int,
2115 'age_limit': 0,
2116 'categories': ['Gaming'],
2117 'tags': 'count:23',
2118 'playable_in_embed': True,
2119 'live_status': 'not_live',
2120 'upload_date': '20220103',
2121 'like_count': int,
2122 'availability': 'public',
2123 'channel': 'Leon Nguyen',
2124 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2125 'channel_follower_count': int
2126 }
2127 }, {
2128 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2129 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2130 'info_dict': {
2131 'id': 'mzZzzBU6lrM',
2132 'ext': 'mp4',
2133 'title': 'I Met GeorgeNotFound In Real Life...',
2134 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2135 'uploader': 'Quackity',
2136 'uploader_id': 'QuackityHQ',
2137 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2138 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2139 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2140 'duration': 955,
2141 'view_count': int,
2142 'age_limit': 0,
2143 'categories': ['Entertainment'],
2144 'tags': 'count:26',
2145 'playable_in_embed': True,
2146 'live_status': 'not_live',
2147 'release_timestamp': 1641172509,
2148 'release_date': '20220103',
2149 'upload_date': '20220103',
2150 'like_count': int,
2151 'availability': 'public',
2152 'channel': 'Quackity',
2153 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2154 'channel_follower_count': int
2155 }
2156 },
2157 { # continuous livestream. Microformat upload date should be preferred.
2158 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2159 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2160 'info_dict': {
2161 'id': 'kgx4WGK0oNU',
2162 '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}',
2163 'ext': 'mp4',
2164 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2165 'availability': 'public',
2166 'age_limit': 0,
2167 'release_timestamp': 1637975704,
2168 'upload_date': '20210619',
2169 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2170 'live_status': 'is_live',
2171 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2172 'uploader': '阿鲍Abao',
2173 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2174 'channel': 'Abao in Tokyo',
2175 'channel_follower_count': int,
2176 'release_date': '20211127',
2177 'tags': 'count:39',
2178 'categories': ['People & Blogs'],
2179 'like_count': int,
2180 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2181 'view_count': int,
2182 'playable_in_embed': True,
2183 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2184 },
2185 'params': {'skip_download': True}
2186 },
2187 ]
2188
2189 @classmethod
2190 def suitable(cls, url):
2191 from ..utils import parse_qs
2192
2193 qs = parse_qs(url)
2194 if qs.get('list', [None])[0]:
2195 return False
2196 return super(YoutubeIE, cls).suitable(url)
2197
2198 def __init__(self, *args, **kwargs):
2199 super(YoutubeIE, self).__init__(*args, **kwargs)
2200 self._code_cache = {}
2201 self._player_cache = {}
2202
2203 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2204 lock = threading.Lock()
2205
2206 is_live = True
2207 start_time = time.time()
2208 formats = [f for f in formats if f.get('is_from_start')]
2209
2210 def refetch_manifest(format_id, delay):
2211 nonlocal formats, start_time, is_live
2212 if time.time() <= start_time + delay:
2213 return
2214
2215 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2216 video_details = traverse_obj(
2217 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2218 microformats = traverse_obj(
2219 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2220 expected_type=dict, default=[])
2221 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2222 start_time = time.time()
2223
2224 def mpd_feed(format_id, delay):
2225 """
2226 @returns (manifest_url, manifest_stream_number, is_live) or None
2227 """
2228 with lock:
2229 refetch_manifest(format_id, delay)
2230
2231 f = next((f for f in formats if f['format_id'] == format_id), None)
2232 if not f:
2233 if not is_live:
2234 self.to_screen(f'{video_id}: Video is no longer live')
2235 else:
2236 self.report_warning(
2237 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2238 return None
2239 return f['manifest_url'], f['manifest_stream_number'], is_live
2240
2241 for f in formats:
2242 f['is_live'] = True
2243 f['protocol'] = 'http_dash_segments_generator'
2244 f['fragments'] = functools.partial(
2245 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2246
2247 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2248 FETCH_SPAN, MAX_DURATION = 5, 432000
2249
2250 mpd_url, stream_number, is_live = None, None, True
2251
2252 begin_index = 0
2253 download_start_time = ctx.get('start') or time.time()
2254
2255 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2256 if lack_early_segments:
2257 self.report_warning(bug_reports_message(
2258 'Starting download from the last 120 hours of the live stream since '
2259 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2260 lack_early_segments = True
2261
2262 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2263 fragments, fragment_base_url = None, None
2264
2265 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2266 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2267 # Obtain from MPD's maximum seq value
2268 old_mpd_url = mpd_url
2269 last_error = ctx.pop('last_error', None)
2270 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2271 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2272 or (mpd_url, stream_number, False))
2273 if not refresh_sequence:
2274 if expire_fast and not is_live:
2275 return False, last_seq
2276 elif old_mpd_url == mpd_url:
2277 return True, last_seq
2278 try:
2279 fmts, _ = self._extract_mpd_formats_and_subtitles(
2280 mpd_url, None, note=False, errnote=False, fatal=False)
2281 except ExtractorError:
2282 fmts = None
2283 if not fmts:
2284 no_fragment_score += 2
2285 return False, last_seq
2286 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2287 fragments = fmt_info['fragments']
2288 fragment_base_url = fmt_info['fragment_base_url']
2289 assert fragment_base_url
2290
2291 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2292 return True, _last_seq
2293
2294 while is_live:
2295 fetch_time = time.time()
2296 if no_fragment_score > 30:
2297 return
2298 if last_segment_url:
2299 # Obtain from "X-Head-Seqnum" header value from each segment
2300 try:
2301 urlh = self._request_webpage(
2302 last_segment_url, None, note=False, errnote=False, fatal=False)
2303 except ExtractorError:
2304 urlh = None
2305 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2306 if last_seq is None:
2307 no_fragment_score += 2
2308 last_segment_url = None
2309 continue
2310 else:
2311 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2312 no_fragment_score += 2
2313 if not should_continue:
2314 continue
2315
2316 if known_idx > last_seq:
2317 last_segment_url = None
2318 continue
2319
2320 last_seq += 1
2321
2322 if begin_index < 0 and known_idx < 0:
2323 # skip from the start when it's negative value
2324 known_idx = last_seq + begin_index
2325 if lack_early_segments:
2326 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2327 try:
2328 for idx in range(known_idx, last_seq):
2329 # do not update sequence here or you'll get skipped some part of it
2330 should_continue, _ = _extract_sequence_from_mpd(False, False)
2331 if not should_continue:
2332 known_idx = idx - 1
2333 raise ExtractorError('breaking out of outer loop')
2334 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2335 yield {
2336 'url': last_segment_url,
2337 }
2338 if known_idx == last_seq:
2339 no_fragment_score += 5
2340 else:
2341 no_fragment_score = 0
2342 known_idx = last_seq
2343 except ExtractorError:
2344 continue
2345
2346 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2347
2348 def _extract_player_url(self, *ytcfgs, webpage=None):
2349 player_url = traverse_obj(
2350 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2351 get_all=False, expected_type=compat_str)
2352 if not player_url:
2353 return
2354 return urljoin('https://www.youtube.com', player_url)
2355
2356 def _download_player_url(self, video_id, fatal=False):
2357 res = self._download_webpage(
2358 'https://www.youtube.com/iframe_api',
2359 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2360 if res:
2361 player_version = self._search_regex(
2362 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2363 if player_version:
2364 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2365
2366 def _signature_cache_id(self, example_sig):
2367 """ Return a string representation of a signature """
2368 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2369
2370 @classmethod
2371 def _extract_player_info(cls, player_url):
2372 for player_re in cls._PLAYER_INFO_RE:
2373 id_m = re.search(player_re, player_url)
2374 if id_m:
2375 break
2376 else:
2377 raise ExtractorError('Cannot identify player %r' % player_url)
2378 return id_m.group('id')
2379
2380 def _load_player(self, video_id, player_url, fatal=True):
2381 player_id = self._extract_player_info(player_url)
2382 if player_id not in self._code_cache:
2383 code = self._download_webpage(
2384 player_url, video_id, fatal=fatal,
2385 note='Downloading player ' + player_id,
2386 errnote='Download of %s failed' % player_url)
2387 if code:
2388 self._code_cache[player_id] = code
2389 return self._code_cache.get(player_id)
2390
2391 def _extract_signature_function(self, video_id, player_url, example_sig):
2392 player_id = self._extract_player_info(player_url)
2393
2394 # Read from filesystem cache
2395 func_id = 'js_%s_%s' % (
2396 player_id, self._signature_cache_id(example_sig))
2397 assert os.path.basename(func_id) == func_id
2398
2399 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2400 if cache_spec is not None:
2401 return lambda s: ''.join(s[i] for i in cache_spec)
2402
2403 code = self._load_player(video_id, player_url)
2404 if code:
2405 res = self._parse_sig_js(code)
2406
2407 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2408 cache_res = res(test_string)
2409 cache_spec = [ord(c) for c in cache_res]
2410
2411 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2412 return res
2413
2414 def _print_sig_code(self, func, example_sig):
2415 if not self.get_param('youtube_print_sig_code'):
2416 return
2417
2418 def gen_sig_code(idxs):
2419 def _genslice(start, end, step):
2420 starts = '' if start == 0 else str(start)
2421 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2422 steps = '' if step == 1 else (':%d' % step)
2423 return 's[%s%s%s]' % (starts, ends, steps)
2424
2425 step = None
2426 # Quelch pyflakes warnings - start will be set when step is set
2427 start = '(Never used)'
2428 for i, prev in zip(idxs[1:], idxs[:-1]):
2429 if step is not None:
2430 if i - prev == step:
2431 continue
2432 yield _genslice(start, prev, step)
2433 step = None
2434 continue
2435 if i - prev in [-1, 1]:
2436 step = i - prev
2437 start = prev
2438 continue
2439 else:
2440 yield 's[%d]' % prev
2441 if step is None:
2442 yield 's[%d]' % i
2443 else:
2444 yield _genslice(start, i, step)
2445
2446 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2447 cache_res = func(test_string)
2448 cache_spec = [ord(c) for c in cache_res]
2449 expr_code = ' + '.join(gen_sig_code(cache_spec))
2450 signature_id_tuple = '(%s)' % (
2451 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2452 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2453 ' return %s\n') % (signature_id_tuple, expr_code)
2454 self.to_screen('Extracted signature function:\n' + code)
2455
2456 def _parse_sig_js(self, jscode):
2457 funcname = self._search_regex(
2458 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2459 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2460 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2461 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2462 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}\(a,\d+\)',
2463 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\(\s*""\s*\)',
2464 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2465 # Obsolete patterns
2466 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2467 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2468 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2469 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2470 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2471 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2472 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2473 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2474 jscode, 'Initial JS player signature function name', group='sig')
2475
2476 jsi = JSInterpreter(jscode)
2477 initial_function = jsi.extract_function(funcname)
2478 return lambda s: initial_function([s])
2479
2480 def _decrypt_signature(self, s, video_id, player_url):
2481 """Turn the encrypted s field into a working signature"""
2482
2483 if player_url is None:
2484 raise ExtractorError('Cannot decrypt signature without player_url')
2485
2486 try:
2487 player_id = (player_url, self._signature_cache_id(s))
2488 if player_id not in self._player_cache:
2489 func = self._extract_signature_function(
2490 video_id, player_url, s
2491 )
2492 self._player_cache[player_id] = func
2493 func = self._player_cache[player_id]
2494 self._print_sig_code(func, s)
2495 return func(s)
2496 except Exception as e:
2497 raise ExtractorError('Signature extraction failed: ' + traceback.format_exc(), cause=e)
2498
2499 def _decrypt_nsig(self, s, video_id, player_url):
2500 """Turn the encrypted n field into a working signature"""
2501 if player_url is None:
2502 raise ExtractorError('Cannot decrypt nsig without player_url')
2503 player_url = urljoin('https://www.youtube.com', player_url)
2504
2505 sig_id = ('nsig_value', s)
2506 if sig_id in self._player_cache:
2507 return self._player_cache[sig_id]
2508
2509 try:
2510 player_id = ('nsig', player_url)
2511 if player_id not in self._player_cache:
2512 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2513 func = self._player_cache[player_id]
2514 self._player_cache[sig_id] = func(s)
2515 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2516 return self._player_cache[sig_id]
2517 except Exception as e:
2518 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2519
2520 def _extract_n_function_name(self, jscode):
2521 nfunc, idx = self._search_regex(
2522 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2523 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2524 if not idx:
2525 return nfunc
2526 return json.loads(js_to_json(self._search_regex(
2527 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2528 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2529
2530 def _extract_n_function(self, video_id, player_url):
2531 player_id = self._extract_player_info(player_url)
2532 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2533
2534 if func_code:
2535 jsi = JSInterpreter(func_code)
2536 else:
2537 jscode = self._load_player(video_id, player_url)
2538 funcname = self._extract_n_function_name(jscode)
2539 jsi = JSInterpreter(jscode)
2540 func_code = jsi.extract_function_code(funcname)
2541 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2542
2543 if self.get_param('youtube_print_sig_code'):
2544 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2545
2546 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2547
2548 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2549 """
2550 Extract signatureTimestamp (sts)
2551 Required to tell API what sig/player version is in use.
2552 """
2553 sts = None
2554 if isinstance(ytcfg, dict):
2555 sts = int_or_none(ytcfg.get('STS'))
2556
2557 if not sts:
2558 # Attempt to extract from player
2559 if player_url is None:
2560 error_msg = 'Cannot extract signature timestamp without player_url.'
2561 if fatal:
2562 raise ExtractorError(error_msg)
2563 self.report_warning(error_msg)
2564 return
2565 code = self._load_player(video_id, player_url, fatal=fatal)
2566 if code:
2567 sts = int_or_none(self._search_regex(
2568 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2569 'JS player signature timestamp', group='sts', fatal=fatal))
2570 return sts
2571
2572 def _mark_watched(self, video_id, player_responses):
2573 playback_url = get_first(
2574 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2575 expected_type=url_or_none)
2576 if not playback_url:
2577 self.report_warning('Unable to mark watched')
2578 return
2579 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2580 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2581
2582 # cpn generation algorithm is reverse engineered from base.js.
2583 # In fact it works even with dummy cpn.
2584 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2585 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
2586
2587 qs.update({
2588 'ver': ['2'],
2589 'cpn': [cpn],
2590 })
2591 playback_url = compat_urlparse.urlunparse(
2592 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2593
2594 self._download_webpage(
2595 playback_url, video_id, 'Marking watched',
2596 'Unable to mark watched', fatal=False)
2597
2598 @staticmethod
2599 def _extract_urls(webpage):
2600 # Embedded YouTube player
2601 entries = [
2602 unescapeHTML(mobj.group('url'))
2603 for mobj in re.finditer(r'''(?x)
2604 (?:
2605 <iframe[^>]+?src=|
2606 data-video-url=|
2607 <embed[^>]+?src=|
2608 embedSWF\(?:\s*|
2609 <object[^>]+data=|
2610 new\s+SWFObject\(
2611 )
2612 (["\'])
2613 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2614 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2615 \1''', webpage)]
2616
2617 # lazyYT YouTube embed
2618 entries.extend(list(map(
2619 unescapeHTML,
2620 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2621
2622 # Wordpress "YouTube Video Importer" plugin
2623 matches = re.findall(r'''(?x)<div[^>]+
2624 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2625 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2626 entries.extend(m[-1] for m in matches)
2627
2628 return entries
2629
2630 @staticmethod
2631 def _extract_url(webpage):
2632 urls = YoutubeIE._extract_urls(webpage)
2633 return urls[0] if urls else None
2634
2635 @classmethod
2636 def extract_id(cls, url):
2637 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2638 if mobj is None:
2639 raise ExtractorError('Invalid URL: %s' % url)
2640 return mobj.group('id')
2641
2642 def _extract_chapters_from_json(self, data, duration):
2643 chapter_list = traverse_obj(
2644 data, (
2645 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2646 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2647 ), expected_type=list)
2648
2649 return self._extract_chapters(
2650 chapter_list,
2651 chapter_time=lambda chapter: float_or_none(
2652 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2653 chapter_title=lambda chapter: traverse_obj(
2654 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2655 duration=duration)
2656
2657 def _extract_chapters_from_engagement_panel(self, data, duration):
2658 content_list = traverse_obj(
2659 data,
2660 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2661 expected_type=list, default=[])
2662 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2663 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2664
2665 return next((
2666 filter(None, (
2667 self._extract_chapters(
2668 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2669 chapter_time, chapter_title, duration)
2670 for contents in content_list
2671 ))), [])
2672
2673 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2674 chapters = []
2675 last_chapter = {'start_time': 0}
2676 for idx, chapter in enumerate(chapter_list or []):
2677 title = chapter_title(chapter)
2678 start_time = chapter_time(chapter)
2679 if start_time is None:
2680 continue
2681 last_chapter['end_time'] = start_time
2682 if start_time < last_chapter['start_time']:
2683 if idx == 1:
2684 chapters.pop()
2685 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2686 else:
2687 self.report_warning(f'Invalid start time for chapter "{title}"')
2688 continue
2689 last_chapter = {'start_time': start_time, 'title': title}
2690 chapters.append(last_chapter)
2691 last_chapter['end_time'] = duration
2692 return chapters
2693
2694 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2695 return self._parse_json(self._search_regex(
2696 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
2697 regex), webpage, name, default='{}'), video_id, fatal=False)
2698
2699 def _extract_comment(self, comment_renderer, parent=None):
2700 comment_id = comment_renderer.get('commentId')
2701 if not comment_id:
2702 return
2703
2704 text = self._get_text(comment_renderer, 'contentText')
2705
2706 # note: timestamp is an estimate calculated from the current time and time_text
2707 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2708 author = self._get_text(comment_renderer, 'authorText')
2709 author_id = try_get(comment_renderer,
2710 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2711
2712 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2713 lambda x: x['likeCount']), compat_str)) or 0
2714 author_thumbnail = try_get(comment_renderer,
2715 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2716
2717 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2718 is_favorited = 'creatorHeart' in (try_get(
2719 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2720 return {
2721 'id': comment_id,
2722 'text': text,
2723 'timestamp': timestamp,
2724 'time_text': time_text,
2725 'like_count': votes,
2726 'is_favorited': is_favorited,
2727 'author': author,
2728 'author_id': author_id,
2729 'author_thumbnail': author_thumbnail,
2730 'author_is_uploader': author_is_uploader,
2731 'parent': parent or 'root'
2732 }
2733
2734 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2735
2736 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2737
2738 def extract_header(contents):
2739 _continuation = None
2740 for content in contents:
2741 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2742 expected_comment_count = self._get_count(
2743 comments_header_renderer, 'countText', 'commentsCount')
2744
2745 if expected_comment_count:
2746 tracker['est_total'] = expected_comment_count
2747 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2748 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2749
2750 sort_menu_item = try_get(
2751 comments_header_renderer,
2752 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2753 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2754
2755 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2756 if not _continuation:
2757 continue
2758
2759 sort_text = str_or_none(sort_menu_item.get('title'))
2760 if not sort_text:
2761 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2762 self.to_screen('Sorting comments by %s' % sort_text.lower())
2763 break
2764 return _continuation
2765
2766 def extract_thread(contents):
2767 if not parent:
2768 tracker['current_page_thread'] = 0
2769 for content in contents:
2770 if not parent and tracker['total_parent_comments'] >= max_parents:
2771 yield
2772 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2773 comment_renderer = get_first(
2774 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2775 expected_type=dict, default={})
2776
2777 comment = self._extract_comment(comment_renderer, parent)
2778 if not comment:
2779 continue
2780
2781 tracker['running_total'] += 1
2782 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2783 yield comment
2784
2785 # Attempt to get the replies
2786 comment_replies_renderer = try_get(
2787 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2788
2789 if comment_replies_renderer:
2790 tracker['current_page_thread'] += 1
2791 comment_entries_iter = self._comment_entries(
2792 comment_replies_renderer, ytcfg, video_id,
2793 parent=comment.get('id'), tracker=tracker)
2794 for reply_comment in itertools.islice(comment_entries_iter, min(max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments']))):
2795 yield reply_comment
2796
2797 # Keeps track of counts across recursive calls
2798 if not tracker:
2799 tracker = dict(
2800 running_total=0,
2801 est_total=0,
2802 current_page_thread=0,
2803 total_parent_comments=0,
2804 total_reply_comments=0)
2805
2806 # TODO: Deprecated
2807 # YouTube comments have a max depth of 2
2808 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2809 if max_depth:
2810 self._downloader.deprecation_warning(
2811 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2812 if max_depth == 1 and parent:
2813 return
2814
2815 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2816 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2817
2818 continuation = self._extract_continuation(root_continuation_data)
2819 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2820 if message and not parent:
2821 self.report_warning(message, video_id=video_id)
2822
2823 response = None
2824 is_first_continuation = parent is None
2825
2826 for page_num in itertools.count(0):
2827 if not continuation:
2828 break
2829 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2830 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2831 if page_num == 0:
2832 if is_first_continuation:
2833 note_prefix = 'Downloading comment section API JSON'
2834 else:
2835 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2836 tracker['current_page_thread'], comment_prog_str)
2837 else:
2838 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2839 ' ' if parent else '', ' replies' if parent else '',
2840 page_num, comment_prog_str)
2841
2842 response = self._extract_response(
2843 item_id=None, query=continuation,
2844 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2845 check_get_keys='onResponseReceivedEndpoints')
2846
2847 continuation_contents = traverse_obj(
2848 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2849
2850 continuation = None
2851 for continuation_section in continuation_contents:
2852 continuation_items = traverse_obj(
2853 continuation_section,
2854 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2855 get_all=False, expected_type=list) or []
2856 if is_first_continuation:
2857 continuation = extract_header(continuation_items)
2858 is_first_continuation = False
2859 if continuation:
2860 break
2861 continue
2862
2863 for entry in extract_thread(continuation_items):
2864 if not entry:
2865 return
2866 yield entry
2867 continuation = self._extract_continuation({'contents': continuation_items})
2868 if continuation:
2869 break
2870
2871 def _get_comments(self, ytcfg, video_id, contents, webpage):
2872 """Entry for comment extraction"""
2873 def _real_comment_extract(contents):
2874 renderer = next((
2875 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2876 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2877 yield from self._comment_entries(renderer, ytcfg, video_id)
2878
2879 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2880 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2881
2882 @staticmethod
2883 def _get_checkok_params():
2884 return {'contentCheckOk': True, 'racyCheckOk': True}
2885
2886 @classmethod
2887 def _generate_player_context(cls, sts=None):
2888 context = {
2889 'html5Preference': 'HTML5_PREF_WANTS',
2890 }
2891 if sts is not None:
2892 context['signatureTimestamp'] = sts
2893 return {
2894 'playbackContext': {
2895 'contentPlaybackContext': context
2896 },
2897 **cls._get_checkok_params()
2898 }
2899
2900 @staticmethod
2901 def _is_agegated(player_response):
2902 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2903 return True
2904
2905 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2906 AGE_GATE_REASONS = (
2907 'confirm your age', 'age-restricted', 'inappropriate', # reason
2908 'age_verification_required', 'age_check_required', # status
2909 )
2910 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2911
2912 @staticmethod
2913 def _is_unplayable(player_response):
2914 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2915
2916 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2917
2918 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2919 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2920 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2921 headers = self.generate_api_headers(
2922 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
2923
2924 yt_query = {'videoId': video_id}
2925 yt_query.update(self._generate_player_context(sts))
2926 return self._extract_response(
2927 item_id=video_id, ep='player', query=yt_query,
2928 ytcfg=player_ytcfg, headers=headers, fatal=True,
2929 default_client=client,
2930 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
2931 ) or None
2932
2933 def _get_requested_clients(self, url, smuggled_data):
2934 requested_clients = []
2935 default = ['android', 'web']
2936 allowed_clients = sorted(
2937 [client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'],
2938 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
2939 for client in self._configuration_arg('player_client'):
2940 if client in allowed_clients:
2941 requested_clients.append(client)
2942 elif client == 'default':
2943 requested_clients.extend(default)
2944 elif client == 'all':
2945 requested_clients.extend(allowed_clients)
2946 else:
2947 self.report_warning(f'Skipping unsupported client {client}')
2948 if not requested_clients:
2949 requested_clients = default
2950
2951 if smuggled_data.get('is_music_url') or self.is_music_url(url):
2952 requested_clients.extend(
2953 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
2954
2955 return orderedSet(requested_clients)
2956
2957 def _extract_player_ytcfg(self, client, video_id):
2958 url = {
2959 'web_music': 'https://music.youtube.com',
2960 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
2961 }.get(client)
2962 if not url:
2963 return {}
2964 webpage = self._download_webpage(url, video_id, fatal=False, note='Downloading %s config' % client.replace('_', ' ').strip())
2965 return self.extract_ytcfg(video_id, webpage) or {}
2966
2967 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
2968 initial_pr = None
2969 if webpage:
2970 initial_pr = self._extract_yt_initial_variable(
2971 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
2972 video_id, 'initial player response')
2973
2974 all_clients = set(clients)
2975 clients = clients[::-1]
2976 prs = []
2977
2978 def append_client(*client_names):
2979 """ Append the first client name that exists but not already used """
2980 for client_name in client_names:
2981 actual_client = _split_innertube_client(client_name)[0]
2982 if actual_client in INNERTUBE_CLIENTS:
2983 if actual_client not in all_clients:
2984 clients.append(client_name)
2985 all_clients.add(actual_client)
2986 return
2987
2988 # Android player_response does not have microFormats which are needed for
2989 # extraction of some data. So we return the initial_pr with formats
2990 # stripped out even if not requested by the user
2991 # See: https://github.com/yt-dlp/yt-dlp/issues/501
2992 if initial_pr:
2993 pr = dict(initial_pr)
2994 pr['streamingData'] = None
2995 prs.append(pr)
2996
2997 last_error = None
2998 tried_iframe_fallback = False
2999 player_url = None
3000 while clients:
3001 client, base_client, variant = _split_innertube_client(clients.pop())
3002 player_ytcfg = master_ytcfg if client == 'web' else {}
3003 if 'configs' not in self._configuration_arg('player_skip'):
3004 player_ytcfg = self._extract_player_ytcfg(client, video_id) or player_ytcfg
3005
3006 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3007 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3008 if 'js' in self._configuration_arg('player_skip'):
3009 require_js_player = False
3010 player_url = None
3011
3012 if not player_url and not tried_iframe_fallback and require_js_player:
3013 player_url = self._download_player_url(video_id)
3014 tried_iframe_fallback = True
3015
3016 try:
3017 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3018 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3019 except ExtractorError as e:
3020 if last_error:
3021 self.report_warning(last_error)
3022 last_error = e
3023 continue
3024
3025 if pr:
3026 prs.append(pr)
3027
3028 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3029 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3030 append_client(f'{base_client}_creator')
3031 elif self._is_agegated(pr):
3032 if variant == 'tv_embedded':
3033 append_client(f'{base_client}_embedded')
3034 elif not variant:
3035 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3036
3037 if last_error:
3038 if not len(prs):
3039 raise last_error
3040 self.report_warning(last_error)
3041 return prs, player_url
3042
3043 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3044 itags, stream_ids = {}, []
3045 itag_qualities, res_qualities = {}, {}
3046 q = qualities([
3047 # Normally tiny is the smallest video-only formats. But
3048 # audio-only formats with unknown quality may get tagged as tiny
3049 'tiny',
3050 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3051 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3052 ])
3053 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3054
3055 for fmt in streaming_formats:
3056 if fmt.get('targetDurationSec'):
3057 continue
3058
3059 itag = str_or_none(fmt.get('itag'))
3060 audio_track = fmt.get('audioTrack') or {}
3061 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3062 if stream_id in stream_ids:
3063 continue
3064
3065 quality = fmt.get('quality')
3066 height = int_or_none(fmt.get('height'))
3067 if quality == 'tiny' or not quality:
3068 quality = fmt.get('audioQuality', '').lower() or quality
3069 # The 3gp format (17) in android client has a quality of "small",
3070 # but is actually worse than other formats
3071 if itag == '17':
3072 quality = 'tiny'
3073 if quality:
3074 if itag:
3075 itag_qualities[itag] = quality
3076 if height:
3077 res_qualities[height] = quality
3078 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3079 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3080 # number of fragment that would subsequently requested with (`&sq=N`)
3081 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3082 continue
3083
3084 fmt_url = fmt.get('url')
3085 if not fmt_url:
3086 sc = compat_parse_qs(fmt.get('signatureCipher'))
3087 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3088 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3089 if not (sc and fmt_url and encrypted_sig):
3090 continue
3091 if not player_url:
3092 continue
3093 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
3094 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
3095 fmt_url += '&' + sp + '=' + signature
3096
3097 query = parse_qs(fmt_url)
3098 throttled = False
3099 if query.get('n'):
3100 try:
3101 fmt_url = update_url_query(fmt_url, {
3102 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3103 except ExtractorError as e:
3104 self.report_warning(
3105 f'nsig extraction failed: You may experience throttling for some formats\n'
3106 f'n = {query["n"][0]} ; player = {player_url}\n{e}', only_once=True)
3107 throttled = True
3108
3109 if itag:
3110 itags[itag] = 'https'
3111 stream_ids.append(stream_id)
3112
3113 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3114 language_preference = (
3115 10 if audio_track.get('audioIsDefault') and 10
3116 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3117 else -1)
3118 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3119 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3120 # Make sure to avoid false positives with small duration differences.
3121 # Eg: __2ABJjxzNo, ySuUZEjARPY
3122 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3123 if is_damaged:
3124 self.report_warning(f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3125 dct = {
3126 'asr': int_or_none(fmt.get('audioSampleRate')),
3127 'filesize': int_or_none(fmt.get('contentLength')),
3128 'format_id': itag,
3129 'format_note': join_nonempty(
3130 '%s%s' % (audio_track.get('displayName') or '',
3131 ' (default)' if language_preference > 0 else ''),
3132 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3133 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3134 'source_preference': -10 if throttled else -1,
3135 'fps': int_or_none(fmt.get('fps')) or None,
3136 'height': height,
3137 'quality': q(quality),
3138 'has_drm': bool(fmt.get('drmFamilies')),
3139 'tbr': tbr,
3140 'url': fmt_url,
3141 'width': int_or_none(fmt.get('width')),
3142 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3143 'desc' if language_preference < -1 else ''),
3144 'language_preference': language_preference,
3145 # Strictly de-prioritize damaged and 3gp formats
3146 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3147 }
3148 mime_mobj = re.match(
3149 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3150 if mime_mobj:
3151 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3152 dct.update(parse_codecs(mime_mobj.group(2)))
3153 no_audio = dct.get('acodec') == 'none'
3154 no_video = dct.get('vcodec') == 'none'
3155 if no_audio:
3156 dct['vbr'] = tbr
3157 if no_video:
3158 dct['abr'] = tbr
3159 if no_audio or no_video:
3160 dct['downloader_options'] = {
3161 # Youtube throttles chunks >~10M
3162 'http_chunk_size': 10485760,
3163 }
3164 if dct.get('ext'):
3165 dct['container'] = dct['ext'] + '_dash'
3166 yield dct
3167
3168 live_from_start = is_live and self.get_param('live_from_start')
3169 skip_manifests = self._configuration_arg('skip')
3170 if not self.get_param('youtube_include_hls_manifest', True):
3171 skip_manifests.append('hls')
3172 get_dash = 'dash' not in skip_manifests and (
3173 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3174 get_hls = not live_from_start and 'hls' not in skip_manifests
3175
3176 def process_manifest_format(f, proto, itag):
3177 if itag in itags:
3178 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3179 return False
3180 itag = f'{itag}-{proto}'
3181 if itag:
3182 f['format_id'] = itag
3183 itags[itag] = proto
3184
3185 f['quality'] = next((
3186 q(qdict[val])
3187 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3188 if val in qdict), -1)
3189 return True
3190
3191 for sd in streaming_data:
3192 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3193 if hls_manifest_url:
3194 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3195 if process_manifest_format(f, 'hls', self._search_regex(
3196 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3197 yield f
3198
3199 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3200 if dash_manifest_url:
3201 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3202 if process_manifest_format(f, 'dash', f['format_id']):
3203 f['filesize'] = int_or_none(self._search_regex(
3204 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3205 if live_from_start:
3206 f['is_from_start'] = True
3207
3208 yield f
3209
3210 def _extract_storyboard(self, player_responses, duration):
3211 spec = get_first(
3212 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3213 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3214 if not base_url:
3215 return
3216 L = len(spec) - 1
3217 for i, args in enumerate(spec):
3218 args = args.split('#')
3219 counts = list(map(int_or_none, args[:5]))
3220 if len(args) != 8 or not all(counts):
3221 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3222 continue
3223 width, height, frame_count, cols, rows = counts
3224 N, sigh = args[6:]
3225
3226 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3227 fragment_count = frame_count / (cols * rows)
3228 fragment_duration = duration / fragment_count
3229 yield {
3230 'format_id': f'sb{i}',
3231 'format_note': 'storyboard',
3232 'ext': 'mhtml',
3233 'protocol': 'mhtml',
3234 'acodec': 'none',
3235 'vcodec': 'none',
3236 'url': url,
3237 'width': width,
3238 'height': height,
3239 'fragments': [{
3240 'url': url.replace('$M', str(j)),
3241 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3242 } for j in range(math.ceil(fragment_count))],
3243 }
3244
3245 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3246 webpage = None
3247 if 'webpage' not in self._configuration_arg('player_skip'):
3248 webpage = self._download_webpage(
3249 webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False)
3250
3251 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3252
3253 player_responses, player_url = self._extract_player_responses(
3254 self._get_requested_clients(url, smuggled_data),
3255 video_id, webpage, master_ytcfg)
3256
3257 return webpage, master_ytcfg, player_responses, player_url
3258
3259 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3260 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3261 is_live = get_first(video_details, 'isLive')
3262 if is_live is None:
3263 is_live = get_first(live_broadcast_details, 'isLiveNow')
3264
3265 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3266 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3267
3268 return live_broadcast_details, is_live, streaming_data, formats
3269
3270 def _real_extract(self, url):
3271 url, smuggled_data = unsmuggle_url(url, {})
3272 video_id = self._match_id(url)
3273
3274 base_url = self.http_scheme() + '//www.youtube.com/'
3275 webpage_url = base_url + 'watch?v=' + video_id
3276
3277 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3278
3279 playability_statuses = traverse_obj(
3280 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3281
3282 trailer_video_id = get_first(
3283 playability_statuses,
3284 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3285 expected_type=str)
3286 if trailer_video_id:
3287 return self.url_result(
3288 trailer_video_id, self.ie_key(), trailer_video_id)
3289
3290 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3291 if webpage else (lambda x: None))
3292
3293 video_details = traverse_obj(
3294 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3295 microformats = traverse_obj(
3296 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3297 expected_type=dict, default=[])
3298 video_title = (
3299 get_first(video_details, 'title')
3300 or self._get_text(microformats, (..., 'title'))
3301 or search_meta(['og:title', 'twitter:title', 'title']))
3302 video_description = get_first(video_details, 'shortDescription')
3303
3304 multifeed_metadata_list = get_first(
3305 player_responses,
3306 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3307 expected_type=str)
3308 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3309 if self.get_param('noplaylist'):
3310 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3311 else:
3312 entries = []
3313 feed_ids = []
3314 for feed in multifeed_metadata_list.split(','):
3315 # Unquote should take place before split on comma (,) since textual
3316 # fields may contain comma as well (see
3317 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3318 feed_data = compat_parse_qs(
3319 compat_urllib_parse_unquote_plus(feed))
3320
3321 def feed_entry(name):
3322 return try_get(
3323 feed_data, lambda x: x[name][0], compat_str)
3324
3325 feed_id = feed_entry('id')
3326 if not feed_id:
3327 continue
3328 feed_title = feed_entry('title')
3329 title = video_title
3330 if feed_title:
3331 title += ' (%s)' % feed_title
3332 entries.append({
3333 '_type': 'url_transparent',
3334 'ie_key': 'Youtube',
3335 'url': smuggle_url(
3336 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3337 {'force_singlefeed': True}),
3338 'title': title,
3339 })
3340 feed_ids.append(feed_id)
3341 self.to_screen(
3342 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3343 % (', '.join(feed_ids), video_id))
3344 return self.playlist_result(
3345 entries, video_id, video_title, video_description)
3346
3347 duration = int_or_none(
3348 get_first(video_details, 'lengthSeconds')
3349 or get_first(microformats, 'lengthSeconds')
3350 or parse_duration(search_meta('duration'))) or None
3351
3352 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3353 video_id, microformats, video_details, player_responses, player_url, duration)
3354
3355 if not formats:
3356 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3357 self.report_drm(video_id)
3358 pemr = get_first(
3359 playability_statuses,
3360 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3361 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3362 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3363 if subreason:
3364 if subreason == 'The uploader has not made this video available in your country.':
3365 countries = get_first(microformats, 'availableCountries')
3366 if not countries:
3367 regions_allowed = search_meta('regionsAllowed')
3368 countries = regions_allowed.split(',') if regions_allowed else None
3369 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3370 reason += f'. {subreason}'
3371 if reason:
3372 self.raise_no_formats(reason, expected=True)
3373
3374 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3375 if not keywords and webpage:
3376 keywords = [
3377 unescapeHTML(m.group('content'))
3378 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3379 for keyword in keywords:
3380 if keyword.startswith('yt:stretch='):
3381 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3382 if mobj:
3383 # NB: float is intentional for forcing float division
3384 w, h = (float(v) for v in mobj.groups())
3385 if w > 0 and h > 0:
3386 ratio = w / h
3387 for f in formats:
3388 if f.get('vcodec') != 'none':
3389 f['stretched_ratio'] = ratio
3390 break
3391 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3392 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3393 if thumbnail_url:
3394 thumbnails.append({
3395 'url': thumbnail_url,
3396 })
3397 original_thumbnails = thumbnails.copy()
3398
3399 # The best resolution thumbnails sometimes does not appear in the webpage
3400 # See: https://github.com/ytdl-org/youtube-dl/issues/29049, https://github.com/yt-dlp/yt-dlp/issues/340
3401 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3402 thumbnail_names = [
3403 'maxresdefault', 'hq720', 'sddefault', 'sd1', 'sd2', 'sd3',
3404 'hqdefault', 'hq1', 'hq2', 'hq3', '0',
3405 'mqdefault', 'mq1', 'mq2', 'mq3',
3406 'default', '1', '2', '3'
3407 ]
3408 n_thumbnail_names = len(thumbnail_names)
3409 thumbnails.extend({
3410 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3411 video_id=video_id, name=name, ext=ext,
3412 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3413 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3414 for thumb in thumbnails:
3415 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3416 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3417 self._remove_duplicate_formats(thumbnails)
3418 self._downloader._sort_thumbnails(original_thumbnails)
3419
3420 category = get_first(microformats, 'category') or search_meta('genre')
3421 channel_id = str_or_none(
3422 get_first(video_details, 'channelId')
3423 or get_first(microformats, 'externalChannelId')
3424 or search_meta('channelId'))
3425 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3426
3427 live_content = get_first(video_details, 'isLiveContent')
3428 is_upcoming = get_first(video_details, 'isUpcoming')
3429 if is_live is None:
3430 if is_upcoming or live_content is False:
3431 is_live = False
3432 if is_upcoming is None and (live_content or is_live):
3433 is_upcoming = False
3434 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3435 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3436 if not duration and live_end_time and live_start_time:
3437 duration = live_end_time - live_start_time
3438
3439 if is_live and self.get_param('live_from_start'):
3440 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3441
3442 formats.extend(self._extract_storyboard(player_responses, duration))
3443
3444 # Source is given priority since formats that throttle are given lower source_preference
3445 # When throttling issue is fully fixed, remove this
3446 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3447
3448 info = {
3449 'id': video_id,
3450 'title': video_title,
3451 'formats': formats,
3452 'thumbnails': thumbnails,
3453 # The best thumbnail that we are sure exists. Prevents unnecessary
3454 # URL checking if user don't care about getting the best possible thumbnail
3455 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3456 'description': video_description,
3457 'uploader': get_first(video_details, 'author'),
3458 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3459 'uploader_url': owner_profile_url,
3460 'channel_id': channel_id,
3461 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3462 'duration': duration,
3463 'view_count': int_or_none(
3464 get_first((video_details, microformats), (..., 'viewCount'))
3465 or search_meta('interactionCount')),
3466 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3467 'age_limit': 18 if (
3468 get_first(microformats, 'isFamilySafe') is False
3469 or search_meta('isFamilyFriendly') == 'false'
3470 or search_meta('og:restrictions:age') == '18+') else 0,
3471 'webpage_url': webpage_url,
3472 'categories': [category] if category else None,
3473 'tags': keywords,
3474 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3475 'is_live': is_live,
3476 'was_live': (False if is_live or is_upcoming or live_content is False
3477 else None if is_live is None or is_upcoming is None
3478 else live_content),
3479 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3480 'release_timestamp': live_start_time,
3481 }
3482
3483 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3484 if pctr:
3485 def get_lang_code(track):
3486 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3487 or track.get('languageCode'))
3488
3489 # Converted into dicts to remove duplicates
3490 captions = {
3491 get_lang_code(sub): sub
3492 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3493 translation_languages = {
3494 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3495 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3496
3497 def process_language(container, base_url, lang_code, sub_name, query):
3498 lang_subs = container.setdefault(lang_code, [])
3499 for fmt in self._SUBTITLE_FORMATS:
3500 query.update({
3501 'fmt': fmt,
3502 })
3503 lang_subs.append({
3504 'ext': fmt,
3505 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3506 'name': sub_name,
3507 })
3508
3509 subtitles, automatic_captions = {}, {}
3510 for lang_code, caption_track in captions.items():
3511 base_url = caption_track.get('baseUrl')
3512 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3513 if not base_url:
3514 continue
3515 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3516 if caption_track.get('kind') != 'asr':
3517 if not lang_code:
3518 continue
3519 process_language(
3520 subtitles, base_url, lang_code, lang_name, {})
3521 if not caption_track.get('isTranslatable'):
3522 continue
3523 for trans_code, trans_name in translation_languages.items():
3524 if not trans_code:
3525 continue
3526 orig_trans_code = trans_code
3527 if caption_track.get('kind') != 'asr':
3528 if 'translated_subs' in self._configuration_arg('skip'):
3529 continue
3530 trans_code += f'-{lang_code}'
3531 trans_name += format_field(lang_name, template=' from %s')
3532 # Add an "-orig" label to the original language so that it can be distinguished.
3533 # The subs are returned without "-orig" as well for compatibility
3534 if lang_code == f'a-{orig_trans_code}':
3535 process_language(
3536 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3537 # Setting tlang=lang returns damaged subtitles.
3538 process_language(automatic_captions, base_url, trans_code, trans_name,
3539 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3540 info['automatic_captions'] = automatic_captions
3541 info['subtitles'] = subtitles
3542
3543 parsed_url = compat_urllib_parse_urlparse(url)
3544 for component in [parsed_url.fragment, parsed_url.query]:
3545 query = compat_parse_qs(component)
3546 for k, v in query.items():
3547 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3548 d_k += '_time'
3549 if d_k not in info and k in s_ks:
3550 info[d_k] = parse_duration(query[k][0])
3551
3552 # Youtube Music Auto-generated description
3553 if video_description:
3554 mobj = re.search(r'(?s)(?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?.+\nAuto-generated by YouTube\.\s*$', video_description)
3555 if mobj:
3556 release_year = mobj.group('release_year')
3557 release_date = mobj.group('release_date')
3558 if release_date:
3559 release_date = release_date.replace('-', '')
3560 if not release_year:
3561 release_year = release_date[:4]
3562 info.update({
3563 'album': mobj.group('album'.strip()),
3564 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3565 'track': mobj.group('track').strip(),
3566 'release_date': release_date,
3567 'release_year': int_or_none(release_year),
3568 })
3569
3570 initial_data = None
3571 if webpage:
3572 initial_data = self._extract_yt_initial_variable(
3573 webpage, self._YT_INITIAL_DATA_RE, video_id,
3574 'yt initial data')
3575 if not initial_data:
3576 query = {'videoId': video_id}
3577 query.update(self._get_checkok_params())
3578 initial_data = self._extract_response(
3579 item_id=video_id, ep='next', fatal=False,
3580 ytcfg=master_ytcfg, query=query,
3581 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3582 note='Downloading initial data API JSON')
3583
3584 try:
3585 # This will error if there is no livechat
3586 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3587 info.setdefault('subtitles', {})['live_chat'] = [{
3588 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
3589 'video_id': video_id,
3590 'ext': 'json',
3591 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3592 }]
3593 except (KeyError, IndexError, TypeError):
3594 pass
3595
3596 if initial_data:
3597 info['chapters'] = (
3598 self._extract_chapters_from_json(initial_data, duration)
3599 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3600 or None)
3601
3602 contents = traverse_obj(
3603 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3604 expected_type=list, default=[])
3605
3606 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3607 if vpir:
3608 stl = vpir.get('superTitleLink')
3609 if stl:
3610 stl = self._get_text(stl)
3611 if try_get(
3612 vpir,
3613 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3614 info['location'] = stl
3615 else:
3616 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
3617 if mobj:
3618 info.update({
3619 'series': mobj.group(1),
3620 'season_number': int(mobj.group(2)),
3621 'episode_number': int(mobj.group(3)),
3622 })
3623 for tlb in (try_get(
3624 vpir,
3625 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3626 list) or []):
3627 tbr = tlb.get('toggleButtonRenderer') or {}
3628 for getter, regex in [(
3629 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3630 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3631 lambda x: x['accessibility'],
3632 lambda x: x['accessibilityData']['accessibilityData'],
3633 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3634 label = (try_get(tbr, getter, dict) or {}).get('label')
3635 if label:
3636 mobj = re.match(regex, label)
3637 if mobj:
3638 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3639 break
3640 sbr_tooltip = try_get(
3641 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3642 if sbr_tooltip:
3643 like_count, dislike_count = sbr_tooltip.split(' / ')
3644 info.update({
3645 'like_count': str_to_int(like_count),
3646 'dislike_count': str_to_int(dislike_count),
3647 })
3648 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3649 if vsir:
3650 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3651 info.update({
3652 'channel': self._get_text(vor, 'title'),
3653 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3654
3655 rows = try_get(
3656 vsir,
3657 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3658 list) or []
3659 multiple_songs = False
3660 for row in rows:
3661 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3662 multiple_songs = True
3663 break
3664 for row in rows:
3665 mrr = row.get('metadataRowRenderer') or {}
3666 mrr_title = mrr.get('title')
3667 if not mrr_title:
3668 continue
3669 mrr_title = self._get_text(mrr, 'title')
3670 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3671 if mrr_title == 'License':
3672 info['license'] = mrr_contents_text
3673 elif not multiple_songs:
3674 if mrr_title == 'Album':
3675 info['album'] = mrr_contents_text
3676 elif mrr_title == 'Artist':
3677 info['artist'] = mrr_contents_text
3678 elif mrr_title == 'Song':
3679 info['track'] = mrr_contents_text
3680
3681 fallbacks = {
3682 'channel': 'uploader',
3683 'channel_id': 'uploader_id',
3684 'channel_url': 'uploader_url',
3685 }
3686
3687 # The upload date for scheduled, live and past live streams / premieres in microformats
3688 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3689 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3690 upload_date = (
3691 unified_strdate(get_first(microformats, 'uploadDate'))
3692 or unified_strdate(search_meta('uploadDate')))
3693 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3694 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d')
3695 info['upload_date'] = upload_date
3696
3697 for to, frm in fallbacks.items():
3698 if not info.get(to):
3699 info[to] = info.get(frm)
3700
3701 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3702 v = info.get(s_k)
3703 if v:
3704 info[d_k] = v
3705
3706 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3707 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3708 is_membersonly = None
3709 is_premium = None
3710 if initial_data and is_private is not None:
3711 is_membersonly = False
3712 is_premium = False
3713 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3714 badge_labels = set()
3715 for content in contents:
3716 if not isinstance(content, dict):
3717 continue
3718 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3719 for badge_label in badge_labels:
3720 if badge_label.lower() == 'members only':
3721 is_membersonly = True
3722 elif badge_label.lower() == 'premium':
3723 is_premium = True
3724 elif badge_label.lower() == 'unlisted':
3725 is_unlisted = True
3726
3727 info['availability'] = self._availability(
3728 is_private=is_private,
3729 needs_premium=is_premium,
3730 needs_subscription=is_membersonly,
3731 needs_auth=info['age_limit'] >= 18,
3732 is_unlisted=None if is_private is None else is_unlisted)
3733
3734 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3735
3736 self.mark_watched(video_id, player_responses)
3737
3738 return info
3739
3740
3741class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3742
3743 @staticmethod
3744 def passthrough_smuggled_data(func):
3745 def _smuggle(entries, smuggled_data):
3746 for entry in entries:
3747 # TODO: Convert URL to music.youtube instead.
3748 # Do we need to passthrough any other smuggled_data?
3749 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3750 yield entry
3751
3752 @functools.wraps(func)
3753 def wrapper(self, url):
3754 url, smuggled_data = unsmuggle_url(url, {})
3755 if self.is_music_url(url):
3756 smuggled_data['is_music_url'] = True
3757 info_dict = func(self, url, smuggled_data)
3758 if smuggled_data and info_dict.get('entries'):
3759 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3760 return info_dict
3761 return wrapper
3762
3763 def _extract_channel_id(self, webpage):
3764 channel_id = self._html_search_meta(
3765 'channelId', webpage, 'channel id', default=None)
3766 if channel_id:
3767 return channel_id
3768 channel_url = self._html_search_meta(
3769 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3770 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3771 'twitter:app:url:googleplay'), webpage, 'channel url')
3772 return self._search_regex(
3773 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3774 channel_url, 'channel id')
3775
3776 @staticmethod
3777 def _extract_basic_item_renderer(item):
3778 # Modified from _extract_grid_item_renderer
3779 known_basic_renderers = (
3780 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3781 )
3782 for key, renderer in item.items():
3783 if not isinstance(renderer, dict):
3784 continue
3785 elif key in known_basic_renderers:
3786 return renderer
3787 elif key.startswith('grid') and key.endswith('Renderer'):
3788 return renderer
3789
3790 def _grid_entries(self, grid_renderer):
3791 for item in grid_renderer['items']:
3792 if not isinstance(item, dict):
3793 continue
3794 renderer = self._extract_basic_item_renderer(item)
3795 if not isinstance(renderer, dict):
3796 continue
3797 title = self._get_text(renderer, 'title')
3798
3799 # playlist
3800 playlist_id = renderer.get('playlistId')
3801 if playlist_id:
3802 yield self.url_result(
3803 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3804 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3805 video_title=title)
3806 continue
3807 # video
3808 video_id = renderer.get('videoId')
3809 if video_id:
3810 yield self._extract_video(renderer)
3811 continue
3812 # channel
3813 channel_id = renderer.get('channelId')
3814 if channel_id:
3815 yield self.url_result(
3816 'https://www.youtube.com/channel/%s' % channel_id,
3817 ie=YoutubeTabIE.ie_key(), video_title=title)
3818 continue
3819 # generic endpoint URL support
3820 ep_url = urljoin('https://www.youtube.com/', try_get(
3821 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3822 compat_str))
3823 if ep_url:
3824 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3825 if ie.suitable(ep_url):
3826 yield self.url_result(
3827 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3828 break
3829
3830 def _music_reponsive_list_entry(self, renderer):
3831 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3832 if video_id:
3833 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3834 ie=YoutubeIE.ie_key(), video_id=video_id)
3835 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3836 if playlist_id:
3837 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3838 if video_id:
3839 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3840 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3841 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3842 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3843 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3844 if browse_id:
3845 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3846 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3847
3848 def _shelf_entries_from_content(self, shelf_renderer):
3849 content = shelf_renderer.get('content')
3850 if not isinstance(content, dict):
3851 return
3852 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3853 if renderer:
3854 # TODO: add support for nested playlists so each shelf is processed
3855 # as separate playlist
3856 # TODO: this includes only first N items
3857 for entry in self._grid_entries(renderer):
3858 yield entry
3859 renderer = content.get('horizontalListRenderer')
3860 if renderer:
3861 # TODO
3862 pass
3863
3864 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3865 ep = try_get(
3866 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3867 compat_str)
3868 shelf_url = urljoin('https://www.youtube.com', ep)
3869 if shelf_url:
3870 # Skipping links to another channels, note that checking for
3871 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3872 # will not work
3873 if skip_channels and '/channels?' in shelf_url:
3874 return
3875 title = self._get_text(shelf_renderer, 'title')
3876 yield self.url_result(shelf_url, video_title=title)
3877 # Shelf may not contain shelf URL, fallback to extraction from content
3878 for entry in self._shelf_entries_from_content(shelf_renderer):
3879 yield entry
3880
3881 def _playlist_entries(self, video_list_renderer):
3882 for content in video_list_renderer['contents']:
3883 if not isinstance(content, dict):
3884 continue
3885 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3886 if not isinstance(renderer, dict):
3887 continue
3888 video_id = renderer.get('videoId')
3889 if not video_id:
3890 continue
3891 yield self._extract_video(renderer)
3892
3893 def _rich_entries(self, rich_grid_renderer):
3894 renderer = try_get(
3895 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3896 video_id = renderer.get('videoId')
3897 if not video_id:
3898 return
3899 yield self._extract_video(renderer)
3900
3901 def _video_entry(self, video_renderer):
3902 video_id = video_renderer.get('videoId')
3903 if video_id:
3904 return self._extract_video(video_renderer)
3905
3906 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3907 url = urljoin('https://youtube.com', traverse_obj(
3908 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3909 if url:
3910 return self.url_result(
3911 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3912
3913 def _post_thread_entries(self, post_thread_renderer):
3914 post_renderer = try_get(
3915 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3916 if not post_renderer:
3917 return
3918 # video attachment
3919 video_renderer = try_get(
3920 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3921 video_id = video_renderer.get('videoId')
3922 if video_id:
3923 entry = self._extract_video(video_renderer)
3924 if entry:
3925 yield entry
3926 # playlist attachment
3927 playlist_id = try_get(
3928 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3929 if playlist_id:
3930 yield self.url_result(
3931 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3932 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3933 # inline video links
3934 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
3935 for run in runs:
3936 if not isinstance(run, dict):
3937 continue
3938 ep_url = try_get(
3939 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
3940 if not ep_url:
3941 continue
3942 if not YoutubeIE.suitable(ep_url):
3943 continue
3944 ep_video_id = YoutubeIE._match_id(ep_url)
3945 if video_id == ep_video_id:
3946 continue
3947 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
3948
3949 def _post_thread_continuation_entries(self, post_thread_continuation):
3950 contents = post_thread_continuation.get('contents')
3951 if not isinstance(contents, list):
3952 return
3953 for content in contents:
3954 renderer = content.get('backstagePostThreadRenderer')
3955 if not isinstance(renderer, dict):
3956 continue
3957 for entry in self._post_thread_entries(renderer):
3958 yield entry
3959
3960 r''' # unused
3961 def _rich_grid_entries(self, contents):
3962 for content in contents:
3963 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
3964 if video_renderer:
3965 entry = self._video_entry(video_renderer)
3966 if entry:
3967 yield entry
3968 '''
3969
3970 def _extract_entries(self, parent_renderer, continuation_list):
3971 # continuation_list is modified in-place with continuation_list = [continuation_token]
3972 continuation_list[:] = [None]
3973 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
3974 for content in contents:
3975 if not isinstance(content, dict):
3976 continue
3977 is_renderer = traverse_obj(
3978 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
3979 expected_type=dict)
3980 if not is_renderer:
3981 renderer = content.get('richItemRenderer')
3982 if renderer:
3983 for entry in self._rich_entries(renderer):
3984 yield entry
3985 continuation_list[0] = self._extract_continuation(parent_renderer)
3986 continue
3987 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
3988 for isr_content in isr_contents:
3989 if not isinstance(isr_content, dict):
3990 continue
3991
3992 known_renderers = {
3993 'playlistVideoListRenderer': self._playlist_entries,
3994 'gridRenderer': self._grid_entries,
3995 'reelShelfRenderer': self._grid_entries,
3996 'shelfRenderer': self._shelf_entries,
3997 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
3998 'backstagePostThreadRenderer': self._post_thread_entries,
3999 'videoRenderer': lambda x: [self._video_entry(x)],
4000 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4001 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4002 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4003 }
4004 for key, renderer in isr_content.items():
4005 if key not in known_renderers:
4006 continue
4007 for entry in known_renderers[key](renderer):
4008 if entry:
4009 yield entry
4010 continuation_list[0] = self._extract_continuation(renderer)
4011 break
4012
4013 if not continuation_list[0]:
4014 continuation_list[0] = self._extract_continuation(is_renderer)
4015
4016 if not continuation_list[0]:
4017 continuation_list[0] = self._extract_continuation(parent_renderer)
4018
4019 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4020 continuation_list = [None]
4021 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4022 tab_content = try_get(tab, lambda x: x['content'], dict)
4023 if not tab_content:
4024 return
4025 parent_renderer = (
4026 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4027 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4028 for entry in extract_entries(parent_renderer):
4029 yield entry
4030 continuation = continuation_list[0]
4031
4032 for page_num in itertools.count(1):
4033 if not continuation:
4034 break
4035 headers = self.generate_api_headers(
4036 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4037 response = self._extract_response(
4038 item_id='%s page %s' % (item_id, page_num),
4039 query=continuation, headers=headers, ytcfg=ytcfg,
4040 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4041
4042 if not response:
4043 break
4044 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4045 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4046 visitor_data = self._extract_visitor_data(response) or visitor_data
4047
4048 known_continuation_renderers = {
4049 'playlistVideoListContinuation': self._playlist_entries,
4050 'gridContinuation': self._grid_entries,
4051 'itemSectionContinuation': self._post_thread_continuation_entries,
4052 'sectionListContinuation': extract_entries, # for feeds
4053 }
4054 continuation_contents = try_get(
4055 response, lambda x: x['continuationContents'], dict) or {}
4056 continuation_renderer = None
4057 for key, value in continuation_contents.items():
4058 if key not in known_continuation_renderers:
4059 continue
4060 continuation_renderer = value
4061 continuation_list = [None]
4062 for entry in known_continuation_renderers[key](continuation_renderer):
4063 yield entry
4064 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4065 break
4066 if continuation_renderer:
4067 continue
4068
4069 known_renderers = {
4070 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4071 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4072 'gridVideoRenderer': (self._grid_entries, 'items'),
4073 'gridChannelRenderer': (self._grid_entries, 'items'),
4074 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4075 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4076 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4077 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4078 }
4079 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4080 continuation_items = try_get(
4081 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4082 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4083 video_items_renderer = None
4084 for key, value in continuation_item.items():
4085 if key not in known_renderers:
4086 continue
4087 video_items_renderer = {known_renderers[key][1]: continuation_items}
4088 continuation_list = [None]
4089 for entry in known_renderers[key][0](video_items_renderer):
4090 yield entry
4091 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4092 break
4093 if video_items_renderer:
4094 continue
4095 break
4096
4097 @staticmethod
4098 def _extract_selected_tab(tabs, fatal=True):
4099 for tab in tabs:
4100 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4101 if renderer.get('selected') is True:
4102 return renderer
4103 else:
4104 if fatal:
4105 raise ExtractorError('Unable to find selected tab')
4106
4107 @classmethod
4108 def _extract_uploader(cls, data):
4109 uploader = {}
4110 renderer = cls._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4111 owner = try_get(
4112 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4113 if owner:
4114 uploader['uploader'] = owner.get('text')
4115 uploader['uploader_id'] = try_get(
4116 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4117 uploader['uploader_url'] = urljoin(
4118 'https://www.youtube.com/',
4119 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4120 return {k: v for k, v in uploader.items() if v is not None}
4121
4122 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4123 playlist_id = title = description = channel_url = channel_name = channel_id = None
4124 tags = []
4125
4126 selected_tab = self._extract_selected_tab(tabs)
4127 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4128 renderer = try_get(
4129 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4130 if renderer:
4131 channel_name = renderer.get('title')
4132 channel_url = renderer.get('channelUrl')
4133 channel_id = renderer.get('externalId')
4134 else:
4135 renderer = try_get(
4136 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4137
4138 if renderer:
4139 title = renderer.get('title')
4140 description = renderer.get('description', '')
4141 playlist_id = channel_id
4142 tags = renderer.get('keywords', '').split()
4143
4144 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4145 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4146 def _get_uncropped(url):
4147 return url_or_none((url or '').split('=')[0] + '=s0')
4148
4149 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4150 if avatar_thumbnails:
4151 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4152 if uncropped_avatar:
4153 avatar_thumbnails.append({
4154 'url': uncropped_avatar,
4155 'id': 'avatar_uncropped',
4156 'preference': 1
4157 })
4158
4159 channel_banners = self._extract_thumbnails(
4160 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4161 for banner in channel_banners:
4162 banner['preference'] = -10
4163
4164 if channel_banners:
4165 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4166 if uncropped_banner:
4167 channel_banners.append({
4168 'url': uncropped_banner,
4169 'id': 'banner_uncropped',
4170 'preference': -5
4171 })
4172
4173 primary_thumbnails = self._extract_thumbnails(
4174 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4175
4176 if playlist_id is None:
4177 playlist_id = item_id
4178
4179 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4180 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4181 if title is None:
4182 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4183 title += format_field(selected_tab, 'title', ' - %s')
4184 title += format_field(selected_tab, 'expandedText', ' - %s')
4185
4186 metadata = {
4187 'playlist_id': playlist_id,
4188 'playlist_title': title,
4189 'playlist_description': description,
4190 'uploader': channel_name,
4191 'uploader_id': channel_id,
4192 'uploader_url': channel_url,
4193 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4194 'tags': tags,
4195 'view_count': self._get_count(playlist_stats, 1),
4196 'availability': self._extract_availability(data),
4197 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4198 'playlist_count': self._get_count(playlist_stats, 0),
4199 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4200 }
4201 if not channel_id:
4202 metadata.update(self._extract_uploader(data))
4203 metadata.update({
4204 'channel': metadata['uploader'],
4205 'channel_id': metadata['uploader_id'],
4206 'channel_url': metadata['uploader_url']})
4207 return self.playlist_result(
4208 self._entries(
4209 selected_tab, playlist_id, ytcfg,
4210 self._extract_account_syncid(ytcfg, data),
4211 self._extract_visitor_data(data, ytcfg)),
4212 **metadata)
4213
4214 def _extract_mix_playlist(self, playlist, playlist_id, data, ytcfg):
4215 first_id = last_id = response = None
4216 for page_num in itertools.count(1):
4217 videos = list(self._playlist_entries(playlist))
4218 if not videos:
4219 return
4220 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4221 if start >= len(videos):
4222 return
4223 for video in videos[start:]:
4224 if video['id'] == first_id:
4225 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
4226 return
4227 yield video
4228 first_id = first_id or videos[0]['id']
4229 last_id = videos[-1]['id']
4230 watch_endpoint = try_get(
4231 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4232 headers = self.generate_api_headers(
4233 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4234 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4235 query = {
4236 'playlistId': playlist_id,
4237 'videoId': watch_endpoint.get('videoId') or last_id,
4238 'index': watch_endpoint.get('index') or len(videos),
4239 'params': watch_endpoint.get('params') or 'OAE%3D'
4240 }
4241 response = self._extract_response(
4242 item_id='%s page %d' % (playlist_id, page_num),
4243 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4244 check_get_keys='contents'
4245 )
4246 playlist = try_get(
4247 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4248
4249 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4250 title = playlist.get('title') or try_get(
4251 data, lambda x: x['titleText']['simpleText'], compat_str)
4252 playlist_id = playlist.get('playlistId') or item_id
4253
4254 # Delegating everything except mix playlists to regular tab-based playlist URL
4255 playlist_url = urljoin(url, try_get(
4256 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4257 compat_str))
4258 if playlist_url and playlist_url != url:
4259 return self.url_result(
4260 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4261 video_title=title)
4262
4263 return self.playlist_result(
4264 self._extract_mix_playlist(playlist, playlist_id, data, ytcfg),
4265 playlist_id=playlist_id, playlist_title=title)
4266
4267 def _extract_availability(self, data):
4268 """
4269 Gets the availability of a given playlist/tab.
4270 Note: Unless YouTube tells us explicitly, we do not assume it is public
4271 @param data: response
4272 """
4273 is_private = is_unlisted = None
4274 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4275 badge_labels = self._extract_badges(renderer)
4276
4277 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4278 privacy_dropdown_entries = try_get(
4279 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4280 for renderer_dict in privacy_dropdown_entries:
4281 is_selected = try_get(
4282 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4283 if not is_selected:
4284 continue
4285 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4286 if label:
4287 badge_labels.add(label.lower())
4288 break
4289
4290 for badge_label in badge_labels:
4291 if badge_label == 'unlisted':
4292 is_unlisted = True
4293 elif badge_label == 'private':
4294 is_private = True
4295 elif badge_label == 'public':
4296 is_unlisted = is_private = False
4297 return self._availability(is_private, False, False, False, is_unlisted)
4298
4299 @staticmethod
4300 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4301 sidebar_renderer = try_get(
4302 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4303 for item in sidebar_renderer:
4304 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4305 if renderer:
4306 return renderer
4307
4308 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4309 """
4310 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4311 """
4312 browse_id = params = None
4313 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4314 if not renderer:
4315 return
4316 menu_renderer = try_get(
4317 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4318 for menu_item in menu_renderer:
4319 if not isinstance(menu_item, dict):
4320 continue
4321 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4322 text = try_get(
4323 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4324 if not text or text.lower() != 'show unavailable videos':
4325 continue
4326 browse_endpoint = try_get(
4327 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4328 browse_id = browse_endpoint.get('browseId')
4329 params = browse_endpoint.get('params')
4330 break
4331
4332 headers = self.generate_api_headers(
4333 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4334 visitor_data=self._extract_visitor_data(data, ytcfg))
4335 query = {
4336 'params': params or 'wgYCCAA=',
4337 'browseId': browse_id or 'VL%s' % item_id
4338 }
4339 return self._extract_response(
4340 item_id=item_id, headers=headers, query=query,
4341 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4342 note='Downloading API JSON with unavailable videos')
4343
4344 def _extract_webpage(self, url, item_id, fatal=True):
4345 retries = self.get_param('extractor_retries', 3)
4346 count = -1
4347 webpage = data = last_error = None
4348 while count < retries:
4349 count += 1
4350 # Sometimes youtube returns a webpage with incomplete ytInitialData
4351 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4352 if last_error:
4353 self.report_warning('%s. Retrying ...' % last_error)
4354 try:
4355 webpage = self._download_webpage(
4356 url, item_id,
4357 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4358 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4359 except ExtractorError as e:
4360 if isinstance(e.cause, network_exceptions):
4361 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4362 last_error = error_to_compat_str(e.cause or e.msg)
4363 if count < retries:
4364 continue
4365 if fatal:
4366 raise
4367 self.report_warning(error_to_compat_str(e))
4368 break
4369 else:
4370 try:
4371 self._extract_and_report_alerts(data)
4372 except ExtractorError as e:
4373 if fatal:
4374 raise
4375 self.report_warning(error_to_compat_str(e))
4376 break
4377
4378 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4379 break
4380
4381 last_error = 'Incomplete yt initial data received'
4382 if count >= retries:
4383 if fatal:
4384 raise ExtractorError(last_error)
4385 self.report_warning(last_error)
4386 break
4387
4388 return webpage, data
4389
4390 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4391 data = None
4392 if 'webpage' not in self._configuration_arg('skip'):
4393 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4394 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4395 # Reject webpage data if redirected to home page without explicitly requesting
4396 selected_tab = self._extract_selected_tab(traverse_obj(
4397 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4398 if (url != 'https://www.youtube.com/feed/recommended'
4399 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4400 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4401 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4402 if fatal:
4403 raise ExtractorError(msg, expected=True)
4404 self.report_warning(msg, only_once=True)
4405 if not data:
4406 if not ytcfg and self.is_authenticated:
4407 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.'
4408 if 'authcheck' not in self._configuration_arg('skip') and fatal:
4409 raise ExtractorError(
4410 msg + ' If you are not downloading private content, or your cookies are only for the first account and channel,'
4411 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4412 expected=True)
4413 self.report_warning(msg, only_once=True)
4414 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4415 return data, ytcfg
4416
4417 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4418 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4419 resolve_response = self._extract_response(
4420 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4421 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4422 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4423 for ep_key, ep in endpoints.items():
4424 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4425 if params:
4426 return self._extract_response(
4427 item_id=item_id, query=params, ep=ep, headers=headers,
4428 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4429 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4430 err_note = 'Failed to resolve url (does the playlist exist?)'
4431 if fatal:
4432 raise ExtractorError(err_note, expected=True)
4433 self.report_warning(err_note, item_id)
4434
4435 _SEARCH_PARAMS = None
4436
4437 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4438 data = {'query': query}
4439 if params is NO_DEFAULT:
4440 params = self._SEARCH_PARAMS
4441 if params:
4442 data['params'] = params
4443
4444 content_keys = (
4445 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4446 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4447 # ytmusic search
4448 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4449 ('continuationContents', ),
4450 )
4451 check_get_keys = tuple(set(keys[0] for keys in content_keys))
4452
4453 continuation_list = [None]
4454 for page_num in itertools.count(1):
4455 data.update(continuation_list[0] or {})
4456 search = self._extract_response(
4457 item_id='query "%s" page %s' % (query, page_num), ep='search', query=data,
4458 default_client=default_client, check_get_keys=check_get_keys)
4459 slr_contents = traverse_obj(search, *content_keys)
4460 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4461 if not continuation_list[0]:
4462 break
4463
4464
4465class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4466 IE_DESC = 'YouTube Tabs'
4467 _VALID_URL = r'''(?x:
4468 https?://
4469 (?:\w+\.)?
4470 (?:
4471 youtube(?:kids)?\.com|
4472 %(invidious)s
4473 )/
4474 (?:
4475 (?P<channel_type>channel|c|user|browse)/|
4476 (?P<not_channel>
4477 feed/|hashtag/|
4478 (?:playlist|watch)\?.*?\blist=
4479 )|
4480 (?!(?:%(reserved_names)s)\b) # Direct URLs
4481 )
4482 (?P<id>[^/?\#&]+)
4483 )''' % {
4484 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4485 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4486 }
4487 IE_NAME = 'youtube:tab'
4488
4489 _TESTS = [{
4490 'note': 'playlists, multipage',
4491 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4492 'playlist_mincount': 94,
4493 'info_dict': {
4494 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4495 'title': 'Igor Kleiner - Playlists',
4496 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4497 'uploader': 'Igor Kleiner',
4498 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4499 'channel': 'Igor Kleiner',
4500 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4501 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4502 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4503 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4504 'channel_follower_count': int
4505 },
4506 }, {
4507 'note': 'playlists, multipage, different order',
4508 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4509 'playlist_mincount': 94,
4510 'info_dict': {
4511 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4512 'title': 'Igor Kleiner - Playlists',
4513 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4514 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4515 'uploader': 'Igor Kleiner',
4516 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4517 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4518 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4519 'channel': 'Igor Kleiner',
4520 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4521 'channel_follower_count': int
4522 },
4523 }, {
4524 'note': 'playlists, series',
4525 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4526 'playlist_mincount': 5,
4527 'info_dict': {
4528 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4529 'title': '3Blue1Brown - Playlists',
4530 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4531 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4532 'uploader': '3Blue1Brown',
4533 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4534 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4535 'channel': '3Blue1Brown',
4536 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4537 'tags': ['Mathematics'],
4538 'channel_follower_count': int
4539 },
4540 }, {
4541 'note': 'playlists, singlepage',
4542 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4543 'playlist_mincount': 4,
4544 'info_dict': {
4545 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4546 'title': 'ThirstForScience - Playlists',
4547 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4548 'uploader': 'ThirstForScience',
4549 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4550 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4551 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4552 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4553 'tags': 'count:13',
4554 'channel': 'ThirstForScience',
4555 'channel_follower_count': int
4556 }
4557 }, {
4558 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4559 'only_matching': True,
4560 }, {
4561 'note': 'basic, single video playlist',
4562 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4563 'info_dict': {
4564 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4565 'uploader': 'Sergey M.',
4566 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4567 'title': 'youtube-dl public playlist',
4568 'description': '',
4569 'tags': [],
4570 'view_count': int,
4571 'modified_date': '20201130',
4572 'channel': 'Sergey M.',
4573 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4574 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4575 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4576 },
4577 'playlist_count': 1,
4578 }, {
4579 'note': 'empty playlist',
4580 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4581 'info_dict': {
4582 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4583 'uploader': 'Sergey M.',
4584 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4585 'title': 'youtube-dl empty playlist',
4586 'tags': [],
4587 'channel': 'Sergey M.',
4588 'description': '',
4589 'modified_date': '20160902',
4590 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4591 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4592 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4593 },
4594 'playlist_count': 0,
4595 }, {
4596 'note': 'Home tab',
4597 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4598 'info_dict': {
4599 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4600 'title': 'lex will - Home',
4601 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4602 'uploader': 'lex will',
4603 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4604 'channel': 'lex will',
4605 'tags': ['bible', 'history', 'prophesy'],
4606 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4607 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4608 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4609 'channel_follower_count': int
4610 },
4611 'playlist_mincount': 2,
4612 }, {
4613 'note': 'Videos tab',
4614 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4615 'info_dict': {
4616 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4617 'title': 'lex will - Videos',
4618 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4619 'uploader': 'lex will',
4620 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4621 'tags': ['bible', 'history', 'prophesy'],
4622 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4623 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4624 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4625 'channel': 'lex will',
4626 'channel_follower_count': int
4627 },
4628 'playlist_mincount': 975,
4629 }, {
4630 'note': 'Videos tab, sorted by popular',
4631 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4632 'info_dict': {
4633 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4634 'title': 'lex will - Videos',
4635 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4636 'uploader': 'lex will',
4637 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4638 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4639 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4640 'channel': 'lex will',
4641 'tags': ['bible', 'history', 'prophesy'],
4642 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4643 'channel_follower_count': int
4644 },
4645 'playlist_mincount': 199,
4646 }, {
4647 'note': 'Playlists tab',
4648 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4649 'info_dict': {
4650 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4651 'title': 'lex will - Playlists',
4652 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4653 'uploader': 'lex will',
4654 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4655 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4656 'channel': 'lex will',
4657 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4658 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4659 'tags': ['bible', 'history', 'prophesy'],
4660 'channel_follower_count': int
4661 },
4662 'playlist_mincount': 17,
4663 }, {
4664 'note': 'Community tab',
4665 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4666 'info_dict': {
4667 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4668 'title': 'lex will - Community',
4669 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4670 'uploader': 'lex will',
4671 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4672 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4673 'channel': 'lex will',
4674 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4675 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4676 'tags': ['bible', 'history', 'prophesy'],
4677 'channel_follower_count': int
4678 },
4679 'playlist_mincount': 18,
4680 }, {
4681 'note': 'Channels tab',
4682 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4683 'info_dict': {
4684 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4685 'title': 'lex will - Channels',
4686 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4687 'uploader': 'lex will',
4688 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4689 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4690 'channel': 'lex will',
4691 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4692 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4693 'tags': ['bible', 'history', 'prophesy'],
4694 'channel_follower_count': int
4695 },
4696 'playlist_mincount': 12,
4697 }, {
4698 'note': 'Search tab',
4699 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4700 'playlist_mincount': 40,
4701 'info_dict': {
4702 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4703 'title': '3Blue1Brown - Search - linear algebra',
4704 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4705 'uploader': '3Blue1Brown',
4706 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4707 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4708 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4709 'tags': ['Mathematics'],
4710 'channel': '3Blue1Brown',
4711 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4712 'channel_follower_count': int
4713 },
4714 }, {
4715 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4716 'only_matching': True,
4717 }, {
4718 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4719 'only_matching': True,
4720 }, {
4721 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4722 'only_matching': True,
4723 }, {
4724 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4725 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4726 'info_dict': {
4727 'title': '29C3: Not my department',
4728 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4729 'uploader': 'Christiaan008',
4730 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4731 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4732 'tags': [],
4733 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4734 'view_count': int,
4735 'modified_date': '20150605',
4736 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4737 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4738 'channel': 'Christiaan008',
4739 },
4740 'playlist_count': 96,
4741 }, {
4742 'note': 'Large playlist',
4743 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4744 'info_dict': {
4745 'title': 'Uploads from Cauchemar',
4746 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4747 'uploader': 'Cauchemar',
4748 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4749 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4750 'tags': [],
4751 'modified_date': r're:\d{8}',
4752 'channel': 'Cauchemar',
4753 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4754 'view_count': int,
4755 'description': '',
4756 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4757 },
4758 'playlist_mincount': 1123,
4759 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4760 }, {
4761 'note': 'even larger playlist, 8832 videos',
4762 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4763 'only_matching': True,
4764 }, {
4765 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4766 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4767 'info_dict': {
4768 'title': 'Uploads from Interstellar Movie',
4769 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4770 'uploader': 'Interstellar Movie',
4771 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4772 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4773 'tags': [],
4774 'view_count': int,
4775 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4776 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4777 'channel': 'Interstellar Movie',
4778 'description': '',
4779 'modified_date': r're:\d{8}',
4780 },
4781 'playlist_mincount': 21,
4782 }, {
4783 'note': 'Playlist with "show unavailable videos" button',
4784 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4785 'info_dict': {
4786 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4787 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4788 'uploader': 'Phim Siêu Nhân Nhật Bản',
4789 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4790 'view_count': int,
4791 'channel': 'Phim Siêu Nhân Nhật Bản',
4792 'tags': [],
4793 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4794 'description': '',
4795 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4796 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4797 'modified_date': r're:\d{8}',
4798 },
4799 'playlist_mincount': 200,
4800 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4801 }, {
4802 'note': 'Playlist with unavailable videos in page 7',
4803 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4804 'info_dict': {
4805 'title': 'Uploads from BlankTV',
4806 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4807 'uploader': 'BlankTV',
4808 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4809 'channel': 'BlankTV',
4810 'channel_url': 'https://www.youtube.com/c/blanktv',
4811 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4812 'view_count': int,
4813 'tags': [],
4814 'uploader_url': 'https://www.youtube.com/c/blanktv',
4815 'modified_date': r're:\d{8}',
4816 'description': '',
4817 },
4818 'playlist_mincount': 1000,
4819 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4820 }, {
4821 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4822 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4823 'info_dict': {
4824 'title': 'Data Analysis with Dr Mike Pound',
4825 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4826 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4827 'uploader': 'Computerphile',
4828 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4829 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4830 'tags': [],
4831 'view_count': int,
4832 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4833 'channel_url': 'https://www.youtube.com/user/Computerphile',
4834 'channel': 'Computerphile',
4835 },
4836 'playlist_mincount': 11,
4837 }, {
4838 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4839 'only_matching': True,
4840 }, {
4841 'note': 'Playlist URL that does not actually serve a playlist',
4842 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4843 'info_dict': {
4844 'id': 'FqZTN594JQw',
4845 'ext': 'webm',
4846 'title': "Smiley's People 01 detective, Adventure Series, Action",
4847 'uploader': 'STREEM',
4848 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4849 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4850 'upload_date': '20150526',
4851 'license': 'Standard YouTube License',
4852 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4853 'categories': ['People & Blogs'],
4854 'tags': list,
4855 'view_count': int,
4856 'like_count': int,
4857 },
4858 'params': {
4859 'skip_download': True,
4860 },
4861 'skip': 'This video is not available.',
4862 'add_ie': [YoutubeIE.ie_key()],
4863 }, {
4864 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4865 'only_matching': True,
4866 }, {
4867 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4868 'only_matching': True,
4869 }, {
4870 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4871 'info_dict': {
4872 'id': 'GgL890LIznQ', # This will keep changing
4873 'ext': 'mp4',
4874 'title': str,
4875 'uploader': 'Sky News',
4876 'uploader_id': 'skynews',
4877 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4878 'upload_date': r're:\d{8}',
4879 'description': str,
4880 'categories': ['News & Politics'],
4881 'tags': list,
4882 'like_count': int,
4883 'release_timestamp': 1642502819,
4884 'channel': 'Sky News',
4885 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4886 'age_limit': 0,
4887 'view_count': int,
4888 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4889 'playable_in_embed': True,
4890 'release_date': '20220118',
4891 'availability': 'public',
4892 'live_status': 'is_live',
4893 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4894 'channel_follower_count': int
4895 },
4896 'params': {
4897 'skip_download': True,
4898 },
4899 'expected_warnings': ['Ignoring subtitle tracks found in '],
4900 }, {
4901 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
4902 'info_dict': {
4903 'id': 'a48o2S1cPoo',
4904 'ext': 'mp4',
4905 'title': 'The Young Turks - Live Main Show',
4906 'uploader': 'The Young Turks',
4907 'uploader_id': 'TheYoungTurks',
4908 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
4909 'upload_date': '20150715',
4910 'license': 'Standard YouTube License',
4911 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
4912 'categories': ['News & Politics'],
4913 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
4914 'like_count': int,
4915 },
4916 'params': {
4917 'skip_download': True,
4918 },
4919 'only_matching': True,
4920 }, {
4921 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
4922 'only_matching': True,
4923 }, {
4924 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
4925 'only_matching': True,
4926 }, {
4927 'note': 'A channel that is not live. Should raise error',
4928 'url': 'https://www.youtube.com/user/numberphile/live',
4929 'only_matching': True,
4930 }, {
4931 'url': 'https://www.youtube.com/feed/trending',
4932 'only_matching': True,
4933 }, {
4934 'url': 'https://www.youtube.com/feed/library',
4935 'only_matching': True,
4936 }, {
4937 'url': 'https://www.youtube.com/feed/history',
4938 'only_matching': True,
4939 }, {
4940 'url': 'https://www.youtube.com/feed/subscriptions',
4941 'only_matching': True,
4942 }, {
4943 'url': 'https://www.youtube.com/feed/watch_later',
4944 'only_matching': True,
4945 }, {
4946 'note': 'Recommended - redirects to home page.',
4947 'url': 'https://www.youtube.com/feed/recommended',
4948 'only_matching': True,
4949 }, {
4950 'note': 'inline playlist with not always working continuations',
4951 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
4952 'only_matching': True,
4953 }, {
4954 'url': 'https://www.youtube.com/course',
4955 'only_matching': True,
4956 }, {
4957 'url': 'https://www.youtube.com/zsecurity',
4958 'only_matching': True,
4959 }, {
4960 'url': 'http://www.youtube.com/NASAgovVideo/videos',
4961 'only_matching': True,
4962 }, {
4963 'url': 'https://www.youtube.com/TheYoungTurks/live',
4964 'only_matching': True,
4965 }, {
4966 'url': 'https://www.youtube.com/hashtag/cctv9',
4967 'info_dict': {
4968 'id': 'cctv9',
4969 'title': '#cctv9',
4970 'tags': [],
4971 },
4972 'playlist_mincount': 350,
4973 }, {
4974 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
4975 'only_matching': True,
4976 }, {
4977 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
4978 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
4979 'only_matching': True
4980 }, {
4981 'note': '/browse/ should redirect to /channel/',
4982 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
4983 'only_matching': True
4984 }, {
4985 'note': 'VLPL, should redirect to playlist?list=PL...',
4986 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
4987 'info_dict': {
4988 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
4989 'uploader': 'NoCopyrightSounds',
4990 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
4991 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
4992 'title': 'NCS Releases',
4993 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
4994 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
4995 'modified_date': r're:\d{8}',
4996 'view_count': int,
4997 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
4998 'tags': [],
4999 'channel': 'NoCopyrightSounds',
5000 },
5001 'playlist_mincount': 166,
5002 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5003 }, {
5004 'note': 'Topic, should redirect to playlist?list=UU...',
5005 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5006 'info_dict': {
5007 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5008 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5009 'title': 'Uploads from Royalty Free Music - Topic',
5010 'uploader': 'Royalty Free Music - Topic',
5011 'tags': [],
5012 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5013 'channel': 'Royalty Free Music - Topic',
5014 'view_count': int,
5015 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5016 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5017 'modified_date': r're:\d{8}',
5018 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5019 'description': '',
5020 },
5021 'expected_warnings': [
5022 'The URL does not have a videos tab',
5023 r'[Uu]navailable videos (are|will be) hidden',
5024 ],
5025 'playlist_mincount': 101,
5026 }, {
5027 'note': 'Topic without a UU playlist',
5028 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5029 'info_dict': {
5030 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5031 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5032 'tags': [],
5033 },
5034 'expected_warnings': [
5035 'the playlist redirect gave error',
5036 ],
5037 'playlist_mincount': 9,
5038 }, {
5039 'note': 'Youtube music Album',
5040 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5041 'info_dict': {
5042 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5043 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5044 'tags': [],
5045 'view_count': int,
5046 'description': '',
5047 'availability': 'unlisted',
5048 'modified_date': r're:\d{8}',
5049 },
5050 'playlist_count': 50,
5051 }, {
5052 'note': 'unlisted single video playlist',
5053 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5054 'info_dict': {
5055 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5056 'uploader': 'colethedj',
5057 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5058 'title': 'yt-dlp unlisted playlist test',
5059 'availability': 'unlisted',
5060 'tags': [],
5061 'modified_date': '20211208',
5062 'channel': 'colethedj',
5063 'view_count': int,
5064 'description': '',
5065 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5066 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5067 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5068 },
5069 'playlist_count': 1,
5070 }, {
5071 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5072 'url': 'https://www.youtube.com/feed/recommended',
5073 'info_dict': {
5074 'id': 'recommended',
5075 'title': 'recommended',
5076 'tags': [],
5077 },
5078 'playlist_mincount': 50,
5079 'params': {
5080 'skip_download': True,
5081 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5082 },
5083 }, {
5084 'note': 'API Fallback: /videos tab, sorted by oldest first',
5085 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5086 'info_dict': {
5087 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5088 'title': 'Cody\'sLab - Videos',
5089 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5090 'uploader': 'Cody\'sLab',
5091 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5092 'channel': 'Cody\'sLab',
5093 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5094 'tags': [],
5095 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5096 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5097 'channel_follower_count': int
5098 },
5099 'playlist_mincount': 650,
5100 'params': {
5101 'skip_download': True,
5102 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5103 },
5104 }, {
5105 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5106 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5107 'info_dict': {
5108 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5109 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5110 'title': 'Uploads from Royalty Free Music - Topic',
5111 'uploader': 'Royalty Free Music - Topic',
5112 'modified_date': r're:\d{8}',
5113 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5114 'description': '',
5115 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5116 'tags': [],
5117 'channel': 'Royalty Free Music - Topic',
5118 'view_count': int,
5119 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5120 },
5121 'expected_warnings': [
5122 'does not have a videos tab',
5123 r'[Uu]navailable videos (are|will be) hidden',
5124 ],
5125 'playlist_mincount': 101,
5126 'params': {
5127 'skip_download': True,
5128 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5129 },
5130 }, {
5131 'note': 'non-standard redirect to regional channel',
5132 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5133 'only_matching': True
5134 }]
5135
5136 @classmethod
5137 def suitable(cls, url):
5138 return False if YoutubeIE.suitable(url) else super(
5139 YoutubeTabIE, cls).suitable(url)
5140
5141 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5142
5143 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5144 def _real_extract(self, url, smuggled_data):
5145 item_id = self._match_id(url)
5146 url = compat_urlparse.urlunparse(
5147 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5148 compat_opts = self.get_param('compat_opts', [])
5149
5150 def get_mobj(url):
5151 mobj = self._URL_RE.match(url).groupdict()
5152 mobj.update((k, '') for k, v in mobj.items() if v is None)
5153 return mobj
5154
5155 mobj, redirect_warning = get_mobj(url), None
5156 # Youtube returns incomplete data if tabname is not lower case
5157 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5158 if is_channel:
5159 if smuggled_data.get('is_music_url'):
5160 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5161 item_id = item_id[2:]
5162 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5163 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5164 mdata = self._extract_tab_endpoint(
5165 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5166 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5167 get_all=False, expected_type=compat_str)
5168 if not murl:
5169 raise ExtractorError('Failed to resolve album to playlist')
5170 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5171 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5172 pre = f'https://www.youtube.com/channel/{item_id}'
5173
5174 original_tab_name = tab
5175 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5176 # Home URLs should redirect to /videos/
5177 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5178 'To download only the videos in the home page, add a "/featured" to the URL')
5179 tab = '/videos'
5180
5181 url = ''.join((pre, tab, post))
5182 mobj = get_mobj(url)
5183
5184 # Handle both video/playlist URLs
5185 qs = parse_qs(url)
5186 video_id, playlist_id = [qs.get(key, [None])[0] for key in ('v', 'list')]
5187
5188 if not video_id and mobj['not_channel'].startswith('watch'):
5189 if not playlist_id:
5190 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5191 raise ExtractorError('Unable to recognize tab page')
5192 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5193 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5194 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5195 mobj = get_mobj(url)
5196
5197 if video_id and playlist_id:
5198 if self.get_param('noplaylist'):
5199 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5200 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5201 ie=YoutubeIE.ie_key(), video_id=video_id)
5202 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5203
5204 data, ytcfg = self._extract_data(url, item_id)
5205
5206 # YouTube may provide a non-standard redirect to the regional channel
5207 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5208 redirect_url = traverse_obj(
5209 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5210 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5211 redirect_url = ''.join((
5212 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5213 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5214 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5215
5216 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5217 if tabs:
5218 selected_tab = self._extract_selected_tab(tabs)
5219 selected_tab_name = selected_tab.get('title', '').lower()
5220 if selected_tab_name == 'home':
5221 selected_tab_name = 'featured'
5222 requested_tab_name = mobj['tab'][1:]
5223 if 'no-youtube-channel-redirect' not in compat_opts:
5224 if requested_tab_name == 'live':
5225 # Live tab should have redirected to the video
5226 raise ExtractorError('The channel is not currently live', expected=True)
5227 if requested_tab_name not in ('', selected_tab_name):
5228 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5229 if not original_tab_name:
5230 if item_id[:2] == 'UC':
5231 # Topic channels don't have /videos. Use the equivalent playlist instead
5232 pl_id = f'UU{item_id[2:]}'
5233 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5234 try:
5235 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5236 except ExtractorError:
5237 redirect_warning += ' and the playlist redirect gave error'
5238 else:
5239 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5240 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5241 if selected_tab_name and selected_tab_name != requested_tab_name:
5242 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5243 else:
5244 raise ExtractorError(redirect_warning, expected=True)
5245
5246 if redirect_warning:
5247 self.to_screen(redirect_warning)
5248 self.write_debug(f'Final URL: {url}')
5249
5250 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5251 if 'no-youtube-unavailable-videos' not in compat_opts:
5252 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5253 self._extract_and_report_alerts(data, only_once=True)
5254 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5255 if tabs:
5256 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5257
5258 playlist = traverse_obj(
5259 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5260 if playlist:
5261 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5262
5263 video_id = traverse_obj(
5264 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5265 if video_id:
5266 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5267 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5268 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5269 ie=YoutubeIE.ie_key(), video_id=video_id)
5270
5271 raise ExtractorError('Unable to recognize tab page')
5272
5273
5274class YoutubePlaylistIE(InfoExtractor):
5275 IE_DESC = 'YouTube playlists'
5276 _VALID_URL = r'''(?x)(?:
5277 (?:https?://)?
5278 (?:\w+\.)?
5279 (?:
5280 (?:
5281 youtube(?:kids)?\.com|
5282 %(invidious)s
5283 )
5284 /.*?\?.*?\blist=
5285 )?
5286 (?P<id>%(playlist_id)s)
5287 )''' % {
5288 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5289 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5290 }
5291 IE_NAME = 'youtube:playlist'
5292 _TESTS = [{
5293 'note': 'issue #673',
5294 'url': 'PLBB231211A4F62143',
5295 'info_dict': {
5296 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5297 'id': 'PLBB231211A4F62143',
5298 'uploader': 'Wickman',
5299 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5300 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5301 'view_count': int,
5302 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5303 'modified_date': r're:\d{8}',
5304 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5305 'channel': 'Wickman',
5306 'tags': [],
5307 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5308 },
5309 'playlist_mincount': 29,
5310 }, {
5311 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5312 'info_dict': {
5313 'title': 'YDL_safe_search',
5314 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5315 },
5316 'playlist_count': 2,
5317 'skip': 'This playlist is private',
5318 }, {
5319 'note': 'embedded',
5320 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5321 'playlist_count': 4,
5322 'info_dict': {
5323 'title': 'JODA15',
5324 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5325 'uploader': 'milan',
5326 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5327 'description': '',
5328 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5329 'tags': [],
5330 'modified_date': '20140919',
5331 'view_count': int,
5332 'channel': 'milan',
5333 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5334 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5335 },
5336 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5337 }, {
5338 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5339 'playlist_mincount': 654,
5340 'info_dict': {
5341 'title': '2018 Chinese New Singles (11/6 updated)',
5342 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5343 'uploader': 'LBK',
5344 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5345 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5346 'channel': 'LBK',
5347 'view_count': int,
5348 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5349 'tags': [],
5350 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5351 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5352 'modified_date': r're:\d{8}',
5353 },
5354 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5355 }, {
5356 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5357 'only_matching': True,
5358 }, {
5359 # music album playlist
5360 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5361 'only_matching': True,
5362 }]
5363
5364 @classmethod
5365 def suitable(cls, url):
5366 if YoutubeTabIE.suitable(url):
5367 return False
5368 from ..utils import parse_qs
5369 qs = parse_qs(url)
5370 if qs.get('v', [None])[0]:
5371 return False
5372 return super(YoutubePlaylistIE, cls).suitable(url)
5373
5374 def _real_extract(self, url):
5375 playlist_id = self._match_id(url)
5376 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5377 url = update_url_query(
5378 'https://www.youtube.com/playlist',
5379 parse_qs(url) or {'list': playlist_id})
5380 if is_music_url:
5381 url = smuggle_url(url, {'is_music_url': True})
5382 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5383
5384
5385class YoutubeYtBeIE(InfoExtractor):
5386 IE_DESC = 'youtu.be'
5387 _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}
5388 _TESTS = [{
5389 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5390 'info_dict': {
5391 'id': 'yeWKywCrFtk',
5392 'ext': 'mp4',
5393 'title': 'Small Scale Baler and Braiding Rugs',
5394 'uploader': 'Backus-Page House Museum',
5395 'uploader_id': 'backuspagemuseum',
5396 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5397 'upload_date': '20161008',
5398 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5399 'categories': ['Nonprofits & Activism'],
5400 'tags': list,
5401 'like_count': int,
5402 'age_limit': 0,
5403 'playable_in_embed': True,
5404 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5405 'channel': 'Backus-Page House Museum',
5406 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5407 'live_status': 'not_live',
5408 'view_count': int,
5409 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5410 'availability': 'public',
5411 'duration': 59,
5412 },
5413 'params': {
5414 'noplaylist': True,
5415 'skip_download': True,
5416 },
5417 }, {
5418 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5419 'only_matching': True,
5420 }]
5421
5422 def _real_extract(self, url):
5423 mobj = self._match_valid_url(url)
5424 video_id = mobj.group('id')
5425 playlist_id = mobj.group('playlist_id')
5426 return self.url_result(
5427 update_url_query('https://www.youtube.com/watch', {
5428 'v': video_id,
5429 'list': playlist_id,
5430 'feature': 'youtu.be',
5431 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5432
5433
5434class YoutubeLivestreamEmbedIE(InfoExtractor):
5435 IE_DESC = 'YouTube livestream embeds'
5436 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5437 _TESTS = [{
5438 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5439 'only_matching': True,
5440 }]
5441
5442 def _real_extract(self, url):
5443 channel_id = self._match_id(url)
5444 return self.url_result(
5445 f'https://www.youtube.com/channel/{channel_id}/live',
5446 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5447
5448
5449class YoutubeYtUserIE(InfoExtractor):
5450 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5451 IE_NAME = 'youtube:user'
5452 _VALID_URL = r'ytuser:(?P<id>.+)'
5453 _TESTS = [{
5454 'url': 'ytuser:phihag',
5455 'only_matching': True,
5456 }]
5457
5458 def _real_extract(self, url):
5459 user_id = self._match_id(url)
5460 return self.url_result(
5461 'https://www.youtube.com/user/%s/videos' % user_id,
5462 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5463
5464
5465class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5466 IE_NAME = 'youtube:favorites'
5467 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5468 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5469 _LOGIN_REQUIRED = True
5470 _TESTS = [{
5471 'url': ':ytfav',
5472 'only_matching': True,
5473 }, {
5474 'url': ':ytfavorites',
5475 'only_matching': True,
5476 }]
5477
5478 def _real_extract(self, url):
5479 return self.url_result(
5480 'https://www.youtube.com/playlist?list=LL',
5481 ie=YoutubeTabIE.ie_key())
5482
5483
5484class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5485 IE_DESC = 'YouTube search'
5486 IE_NAME = 'youtube:search'
5487 _SEARCH_KEY = 'ytsearch'
5488 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5489 _TESTS = [{
5490 'url': 'ytsearch5:youtube-dl test video',
5491 'playlist_count': 5,
5492 'info_dict': {
5493 'id': 'youtube-dl test video',
5494 'title': 'youtube-dl test video',
5495 }
5496 }]
5497
5498
5499class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5500 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5501 _SEARCH_KEY = 'ytsearchdate'
5502 IE_DESC = 'YouTube search, newest videos first'
5503 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5504 _TESTS = [{
5505 'url': 'ytsearchdate5:youtube-dl test video',
5506 'playlist_count': 5,
5507 'info_dict': {
5508 'id': 'youtube-dl test video',
5509 'title': 'youtube-dl test video',
5510 }
5511 }]
5512
5513
5514class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5515 IE_DESC = 'YouTube search URLs with sorting and filter support'
5516 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5517 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5518 _TESTS = [{
5519 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5520 'playlist_mincount': 5,
5521 'info_dict': {
5522 'id': 'youtube-dl test video',
5523 'title': 'youtube-dl test video',
5524 }
5525 }, {
5526 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5527 'playlist_mincount': 5,
5528 'info_dict': {
5529 'id': 'python',
5530 'title': 'python',
5531 }
5532 }, {
5533 'url': 'https://www.youtube.com/results?search_query=%23cats',
5534 'playlist_mincount': 1,
5535 'info_dict': {
5536 'id': '#cats',
5537 'title': '#cats',
5538 'entries': [{
5539 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5540 'title': '#cats',
5541 }],
5542 },
5543 }, {
5544 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5545 'only_matching': True,
5546 }]
5547
5548 def _real_extract(self, url):
5549 qs = parse_qs(url)
5550 query = (qs.get('search_query') or qs.get('q'))[0]
5551 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5552
5553
5554class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5555 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5556 IE_NAME = 'youtube:music:search_url'
5557 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5558 _TESTS = [{
5559 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5560 'playlist_count': 16,
5561 'info_dict': {
5562 'id': 'royalty free music',
5563 'title': 'royalty free music',
5564 }
5565 }, {
5566 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5567 'playlist_mincount': 30,
5568 'info_dict': {
5569 'id': 'royalty free music - songs',
5570 'title': 'royalty free music - songs',
5571 },
5572 'params': {'extract_flat': 'in_playlist'}
5573 }, {
5574 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5575 'playlist_mincount': 30,
5576 'info_dict': {
5577 'id': 'royalty free music - community playlists',
5578 'title': 'royalty free music - community playlists',
5579 },
5580 'params': {'extract_flat': 'in_playlist'}
5581 }]
5582
5583 _SECTIONS = {
5584 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5585 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5586 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5587 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5588 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5589 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5590 }
5591
5592 def _real_extract(self, url):
5593 qs = parse_qs(url)
5594 query = (qs.get('search_query') or qs.get('q'))[0]
5595 params = qs.get('sp', (None,))[0]
5596 if params:
5597 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5598 else:
5599 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5600 params = self._SECTIONS.get(section)
5601 if not params:
5602 section = None
5603 title = join_nonempty(query, section, delim=' - ')
5604 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5605
5606
5607class YoutubeFeedsInfoExtractor(InfoExtractor):
5608 """
5609 Base class for feed extractors
5610 Subclasses must define the _FEED_NAME property.
5611 """
5612 _LOGIN_REQUIRED = True
5613 _TESTS = []
5614
5615 @property
5616 def IE_NAME(self):
5617 return 'youtube:%s' % self._FEED_NAME
5618
5619 def _real_extract(self, url):
5620 return self.url_result(
5621 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5622
5623
5624class YoutubeWatchLaterIE(InfoExtractor):
5625 IE_NAME = 'youtube:watchlater'
5626 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5627 _VALID_URL = r':ytwatchlater'
5628 _TESTS = [{
5629 'url': ':ytwatchlater',
5630 'only_matching': True,
5631 }]
5632
5633 def _real_extract(self, url):
5634 return self.url_result(
5635 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5636
5637
5638class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5639 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5640 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5641 _FEED_NAME = 'recommended'
5642 _LOGIN_REQUIRED = False
5643 _TESTS = [{
5644 'url': ':ytrec',
5645 'only_matching': True,
5646 }, {
5647 'url': ':ytrecommended',
5648 'only_matching': True,
5649 }, {
5650 'url': 'https://youtube.com',
5651 'only_matching': True,
5652 }]
5653
5654
5655class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5656 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5657 _VALID_URL = r':ytsub(?:scription)?s?'
5658 _FEED_NAME = 'subscriptions'
5659 _TESTS = [{
5660 'url': ':ytsubs',
5661 'only_matching': True,
5662 }, {
5663 'url': ':ytsubscriptions',
5664 'only_matching': True,
5665 }]
5666
5667
5668class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5669 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5670 _VALID_URL = r':ythis(?:tory)?'
5671 _FEED_NAME = 'history'
5672 _TESTS = [{
5673 'url': ':ythistory',
5674 'only_matching': True,
5675 }]
5676
5677
5678class YoutubeTruncatedURLIE(InfoExtractor):
5679 IE_NAME = 'youtube:truncated_url'
5680 IE_DESC = False # Do not list
5681 _VALID_URL = r'''(?x)
5682 (?:https?://)?
5683 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5684 (?:watch\?(?:
5685 feature=[a-z_]+|
5686 annotation_id=annotation_[^&]+|
5687 x-yt-cl=[0-9]+|
5688 hl=[^&]*|
5689 t=[0-9]+
5690 )?
5691 |
5692 attribution_link\?a=[^&]+
5693 )
5694 $
5695 '''
5696
5697 _TESTS = [{
5698 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5699 'only_matching': True,
5700 }, {
5701 'url': 'https://www.youtube.com/watch?',
5702 'only_matching': True,
5703 }, {
5704 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5705 'only_matching': True,
5706 }, {
5707 'url': 'https://www.youtube.com/watch?feature=foo',
5708 'only_matching': True,
5709 }, {
5710 'url': 'https://www.youtube.com/watch?hl=en-GB',
5711 'only_matching': True,
5712 }, {
5713 'url': 'https://www.youtube.com/watch?t=2372',
5714 'only_matching': True,
5715 }]
5716
5717 def _real_extract(self, url):
5718 raise ExtractorError(
5719 'Did you forget to quote the URL? Remember that & is a meta '
5720 'character in most shells, so you want to put the URL in quotes, '
5721 'like youtube-dl '
5722 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5723 ' or simply youtube-dl BaW_jenozKc .',
5724 expected=True)
5725
5726
5727class YoutubeClipIE(InfoExtractor):
5728 IE_NAME = 'youtube:clip'
5729 IE_DESC = False # Do not list
5730 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5731
5732 def _real_extract(self, url):
5733 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5734 return self.url_result(url, 'Generic')
5735
5736
5737class YoutubeTruncatedIDIE(InfoExtractor):
5738 IE_NAME = 'youtube:truncated_id'
5739 IE_DESC = False # Do not list
5740 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5741
5742 _TESTS = [{
5743 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5744 'only_matching': True,
5745 }]
5746
5747 def _real_extract(self, url):
5748 video_id = self._match_id(url)
5749 raise ExtractorError(
5750 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
5751 expected=True)