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