]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[extractor/playsuisse] Add extractor (#845)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import copy
4 import datetime
5 import hashlib
6 import itertools
7 import json
8 import math
9 import os.path
10 import random
11 import re
12 import sys
13 import threading
14 import time
15 import traceback
16
17 from .common import InfoExtractor, SearchInfoExtractor
18 from ..compat import functools
19 from ..compat import (
20 compat_chr,
21 compat_HTTPError,
22 compat_parse_qs,
23 compat_str,
24 compat_urllib_parse_unquote_plus,
25 compat_urllib_parse_urlencode,
26 compat_urllib_parse_urlparse,
27 compat_urlparse,
28 )
29 from ..jsinterp import JSInterpreter
30 from ..utils import (
31 NO_DEFAULT,
32 ExtractorError,
33 bug_reports_message,
34 classproperty,
35 clean_html,
36 datetime_from_str,
37 dict_get,
38 error_to_compat_str,
39 float_or_none,
40 format_field,
41 get_first,
42 int_or_none,
43 is_html,
44 join_nonempty,
45 js_to_json,
46 mimetype2ext,
47 network_exceptions,
48 orderedSet,
49 parse_codecs,
50 parse_count,
51 parse_duration,
52 parse_iso8601,
53 parse_qs,
54 qualities,
55 remove_end,
56 remove_start,
57 smuggle_url,
58 str_or_none,
59 str_to_int,
60 strftime_or_none,
61 traverse_obj,
62 try_get,
63 unescapeHTML,
64 unified_strdate,
65 unified_timestamp,
66 unsmuggle_url,
67 update_url_query,
68 url_or_none,
69 urljoin,
70 variadic,
71 )
72
73 # any clients starting with _ cannot be explicity requested by the user
74 INNERTUBE_CLIENTS = {
75 'web': {
76 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
77 'INNERTUBE_CONTEXT': {
78 'client': {
79 'clientName': 'WEB',
80 'clientVersion': '2.20211221.00.00',
81 }
82 },
83 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
84 },
85 'web_embedded': {
86 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
87 'INNERTUBE_CONTEXT': {
88 'client': {
89 'clientName': 'WEB_EMBEDDED_PLAYER',
90 'clientVersion': '1.20211215.00.01',
91 },
92 },
93 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
94 },
95 'web_music': {
96 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
97 'INNERTUBE_HOST': 'music.youtube.com',
98 'INNERTUBE_CONTEXT': {
99 'client': {
100 'clientName': 'WEB_REMIX',
101 'clientVersion': '1.20211213.00.00',
102 }
103 },
104 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
105 },
106 'web_creator': {
107 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
108 'INNERTUBE_CONTEXT': {
109 'client': {
110 'clientName': 'WEB_CREATOR',
111 'clientVersion': '1.20211220.02.00',
112 }
113 },
114 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
115 },
116 'android': {
117 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
118 'INNERTUBE_CONTEXT': {
119 'client': {
120 'clientName': 'ANDROID',
121 'clientVersion': '16.49',
122 }
123 },
124 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
125 'REQUIRE_JS_PLAYER': False
126 },
127 'android_embedded': {
128 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
129 'INNERTUBE_CONTEXT': {
130 'client': {
131 'clientName': 'ANDROID_EMBEDDED_PLAYER',
132 'clientVersion': '16.49',
133 },
134 },
135 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
136 'REQUIRE_JS_PLAYER': False
137 },
138 'android_music': {
139 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
140 'INNERTUBE_CONTEXT': {
141 'client': {
142 'clientName': 'ANDROID_MUSIC',
143 'clientVersion': '4.57',
144 }
145 },
146 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
147 'REQUIRE_JS_PLAYER': False
148 },
149 'android_creator': {
150 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
151 'INNERTUBE_CONTEXT': {
152 'client': {
153 'clientName': 'ANDROID_CREATOR',
154 'clientVersion': '21.47',
155 },
156 },
157 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
158 'REQUIRE_JS_PLAYER': False
159 },
160 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
161 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
162 'ios': {
163 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
164 'INNERTUBE_CONTEXT': {
165 'client': {
166 'clientName': 'IOS',
167 'clientVersion': '16.46',
168 'deviceModel': 'iPhone14,3',
169 }
170 },
171 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
172 'REQUIRE_JS_PLAYER': False
173 },
174 'ios_embedded': {
175 'INNERTUBE_CONTEXT': {
176 'client': {
177 'clientName': 'IOS_MESSAGES_EXTENSION',
178 'clientVersion': '16.46',
179 'deviceModel': 'iPhone14,3',
180 },
181 },
182 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
183 'REQUIRE_JS_PLAYER': False
184 },
185 'ios_music': {
186 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
187 'INNERTUBE_CONTEXT': {
188 'client': {
189 'clientName': 'IOS_MUSIC',
190 'clientVersion': '4.57',
191 },
192 },
193 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
194 'REQUIRE_JS_PLAYER': False
195 },
196 'ios_creator': {
197 'INNERTUBE_CONTEXT': {
198 'client': {
199 'clientName': 'IOS_CREATOR',
200 'clientVersion': '21.47',
201 },
202 },
203 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
204 'REQUIRE_JS_PLAYER': False
205 },
206 # mweb has 'ultralow' formats
207 # See: https://github.com/yt-dlp/yt-dlp/pull/557
208 'mweb': {
209 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
210 'INNERTUBE_CONTEXT': {
211 'client': {
212 'clientName': 'MWEB',
213 'clientVersion': '2.20211221.01.00',
214 }
215 },
216 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
217 },
218 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
219 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
220 'tv_embedded': {
221 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
222 'INNERTUBE_CONTEXT': {
223 'client': {
224 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
225 'clientVersion': '2.0',
226 },
227 },
228 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
229 },
230 }
231
232
233 def _split_innertube_client(client_name):
234 variant, *base = client_name.rsplit('.', 1)
235 if base:
236 return variant, base[0], variant
237 base, *variant = client_name.split('_', 1)
238 return client_name, base, variant[0] if variant else None
239
240
241 def build_innertube_clients():
242 THIRD_PARTY = {
243 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
244 }
245 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
246 priority = qualities(BASE_CLIENTS[::-1])
247
248 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
249 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
250 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
251 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
252 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
253
254 _, base_client, variant = _split_innertube_client(client)
255 ytcfg['priority'] = 10 * priority(base_client)
256
257 if not variant:
258 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
259 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
260 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
261 embedscreen['priority'] -= 3
262 elif variant == 'embedded':
263 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
264 ytcfg['priority'] -= 2
265 else:
266 ytcfg['priority'] -= 3
267
268
269 build_innertube_clients()
270
271
272 class YoutubeBaseInfoExtractor(InfoExtractor):
273 """Provide base functions for Youtube extractors"""
274
275 _RESERVED_NAMES = (
276 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
277 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
278 r'browse|oembed|get_video_info|iframe_api|s/player|'
279 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
280
281 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
282
283 # _NETRC_MACHINE = 'youtube'
284
285 # If True it will raise an error if no login info is provided
286 _LOGIN_REQUIRED = False
287
288 _INVIDIOUS_SITES = (
289 # invidious-redirect websites
290 r'(?:www\.)?redirect\.invidious\.io',
291 r'(?:(?:www|dev)\.)?invidio\.us',
292 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
293 r'(?:www\.)?invidious\.pussthecat\.org',
294 r'(?:www\.)?invidious\.zee\.li',
295 r'(?:www\.)?invidious\.ethibox\.fr',
296 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
297 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
298 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
299 # youtube-dl invidious instances list
300 r'(?:(?:www|no)\.)?invidiou\.sh',
301 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
302 r'(?:www\.)?invidious\.kabi\.tk',
303 r'(?:www\.)?invidious\.mastodon\.host',
304 r'(?:www\.)?invidious\.zapashcanon\.fr',
305 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
306 r'(?:www\.)?invidious\.tinfoil-hat\.net',
307 r'(?:www\.)?invidious\.himiko\.cloud',
308 r'(?:www\.)?invidious\.reallyancient\.tech',
309 r'(?:www\.)?invidious\.tube',
310 r'(?:www\.)?invidiou\.site',
311 r'(?:www\.)?invidious\.site',
312 r'(?:www\.)?invidious\.xyz',
313 r'(?:www\.)?invidious\.nixnet\.xyz',
314 r'(?:www\.)?invidious\.048596\.xyz',
315 r'(?:www\.)?invidious\.drycat\.fr',
316 r'(?:www\.)?inv\.skyn3t\.in',
317 r'(?:www\.)?tube\.poal\.co',
318 r'(?:www\.)?tube\.connect\.cafe',
319 r'(?:www\.)?vid\.wxzm\.sx',
320 r'(?:www\.)?vid\.mint\.lgbt',
321 r'(?:www\.)?vid\.puffyan\.us',
322 r'(?:www\.)?yewtu\.be',
323 r'(?:www\.)?yt\.elukerio\.org',
324 r'(?:www\.)?yt\.lelux\.fi',
325 r'(?:www\.)?invidious\.ggc-project\.de',
326 r'(?:www\.)?yt\.maisputain\.ovh',
327 r'(?:www\.)?ytprivate\.com',
328 r'(?:www\.)?invidious\.13ad\.de',
329 r'(?:www\.)?invidious\.toot\.koeln',
330 r'(?:www\.)?invidious\.fdn\.fr',
331 r'(?:www\.)?watch\.nettohikari\.com',
332 r'(?:www\.)?invidious\.namazso\.eu',
333 r'(?:www\.)?invidious\.silkky\.cloud',
334 r'(?:www\.)?invidious\.exonip\.de',
335 r'(?:www\.)?invidious\.riverside\.rocks',
336 r'(?:www\.)?invidious\.blamefran\.net',
337 r'(?:www\.)?invidious\.moomoo\.de',
338 r'(?:www\.)?ytb\.trom\.tf',
339 r'(?:www\.)?yt\.cyberhost\.uk',
340 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
341 r'(?:www\.)?qklhadlycap4cnod\.onion',
342 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
343 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
344 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
345 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
346 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
347 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
348 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
349 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
350 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
351 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
352 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
353 r'(?:www\.)?piped\.kavin\.rocks',
354 r'(?:www\.)?piped\.silkky\.cloud',
355 r'(?:www\.)?piped\.tokhmi\.xyz',
356 r'(?:www\.)?piped\.moomoo\.me',
357 r'(?:www\.)?il\.ax',
358 r'(?:www\.)?piped\.syncpundit\.com',
359 r'(?:www\.)?piped\.mha\.fi',
360 r'(?:www\.)?piped\.mint\.lgbt',
361 r'(?:www\.)?piped\.privacy\.com\.de',
362 )
363
364 def _initialize_consent(self):
365 cookies = self._get_cookies('https://www.youtube.com/')
366 if cookies.get('__Secure-3PSID'):
367 return
368 consent_id = None
369 consent = cookies.get('CONSENT')
370 if consent:
371 if 'YES' in consent.value:
372 return
373 consent_id = self._search_regex(
374 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
375 if not consent_id:
376 consent_id = random.randint(100, 999)
377 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
378
379 def _initialize_pref(self):
380 cookies = self._get_cookies('https://www.youtube.com/')
381 pref_cookie = cookies.get('PREF')
382 pref = {}
383 if pref_cookie:
384 try:
385 pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
386 except ValueError:
387 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
388 pref.update({'hl': 'en', 'tz': 'UTC'})
389 self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
390
391 def _real_initialize(self):
392 self._initialize_pref()
393 self._initialize_consent()
394 self._check_login_required()
395
396 def _check_login_required(self):
397 if self._LOGIN_REQUIRED and not self._cookies_passed:
398 self.raise_login_required('Login details are needed to download this content', method='cookies')
399
400 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
401 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
402 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
403
404 def _get_default_ytcfg(self, client='web'):
405 return copy.deepcopy(INNERTUBE_CLIENTS[client])
406
407 def _get_innertube_host(self, client='web'):
408 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
409
410 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
411 # try_get but with fallback to default ytcfg client values when present
412 _func = lambda y: try_get(y, getter, expected_type)
413 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
414
415 def _extract_client_name(self, ytcfg, default_client='web'):
416 return self._ytcfg_get_safe(
417 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
418 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
419
420 def _extract_client_version(self, ytcfg, default_client='web'):
421 return self._ytcfg_get_safe(
422 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
423 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
424
425 def _extract_api_key(self, ytcfg=None, default_client='web'):
426 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
427
428 def _extract_context(self, ytcfg=None, default_client='web'):
429 context = get_first(
430 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
431 # Enforce language and tz for extraction
432 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
433 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
434 return context
435
436 _SAPISID = None
437
438 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
439 time_now = round(time.time())
440 if self._SAPISID is None:
441 yt_cookies = self._get_cookies('https://www.youtube.com')
442 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
443 # See: https://github.com/yt-dlp/yt-dlp/issues/393
444 sapisid_cookie = dict_get(
445 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
446 if sapisid_cookie and sapisid_cookie.value:
447 self._SAPISID = sapisid_cookie.value
448 self.write_debug('Extracted SAPISID cookie')
449 # SAPISID cookie is required if not already present
450 if not yt_cookies.get('SAPISID'):
451 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
452 self._set_cookie(
453 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
454 else:
455 self._SAPISID = False
456 if not self._SAPISID:
457 return None
458 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
459 sapisidhash = hashlib.sha1(
460 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
461 return f'SAPISIDHASH {time_now}_{sapisidhash}'
462
463 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
464 note='Downloading API JSON', errnote='Unable to download API page',
465 context=None, api_key=None, api_hostname=None, default_client='web'):
466
467 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
468 data.update(query)
469 real_headers = self.generate_api_headers(default_client=default_client)
470 real_headers.update({'content-type': 'application/json'})
471 if headers:
472 real_headers.update(headers)
473 return self._download_json(
474 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
475 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
476 data=json.dumps(data).encode('utf8'), headers=real_headers,
477 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
478
479 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
480 data = self._search_regex(
481 (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
482 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
483 if data:
484 return self._parse_json(data, item_id, fatal=fatal)
485
486 @staticmethod
487 def _extract_session_index(*data):
488 """
489 Index of current account in account list.
490 See: https://github.com/yt-dlp/yt-dlp/pull/519
491 """
492 for ytcfg in data:
493 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
494 if session_index is not None:
495 return session_index
496
497 # Deprecated?
498 def _extract_identity_token(self, ytcfg=None, webpage=None):
499 if ytcfg:
500 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
501 if token:
502 return token
503 if webpage:
504 return self._search_regex(
505 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
506 'identity token', default=None, fatal=False)
507
508 @staticmethod
509 def _extract_account_syncid(*args):
510 """
511 Extract syncId required to download private playlists of secondary channels
512 @params response and/or ytcfg
513 """
514 for data in args:
515 # ytcfg includes channel_syncid if on secondary channel
516 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
517 if delegated_sid:
518 return delegated_sid
519 sync_ids = (try_get(
520 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
521 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
522 if len(sync_ids) >= 2 and sync_ids[1]:
523 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
524 # and just "user_syncid||" for primary channel. We only want the channel_syncid
525 return sync_ids[0]
526
527 @staticmethod
528 def _extract_visitor_data(*args):
529 """
530 Extracts visitorData from an API response or ytcfg
531 Appears to be used to track session state
532 """
533 return get_first(
534 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
535 expected_type=str)
536
537 @functools.cached_property
538 def is_authenticated(self):
539 return bool(self._generate_sapisidhash_header())
540
541 def extract_ytcfg(self, video_id, webpage):
542 if not webpage:
543 return {}
544 return self._parse_json(
545 self._search_regex(
546 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
547 default='{}'), video_id, fatal=False) or {}
548
549 def generate_api_headers(
550 self, *, ytcfg=None, account_syncid=None, session_index=None,
551 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
552
553 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
554 headers = {
555 'X-YouTube-Client-Name': compat_str(
556 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
557 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
558 'Origin': origin,
559 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
560 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
561 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
562 }
563 if session_index is None:
564 session_index = self._extract_session_index(ytcfg)
565 if account_syncid or session_index is not None:
566 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
567
568 auth = self._generate_sapisidhash_header(origin)
569 if auth is not None:
570 headers['Authorization'] = auth
571 headers['X-Origin'] = origin
572 return {h: v for h, v in headers.items() if v is not None}
573
574 def _download_ytcfg(self, client, video_id):
575 url = {
576 'web': 'https://www.youtube.com',
577 'web_music': 'https://music.youtube.com',
578 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
579 }.get(client)
580 if not url:
581 return {}
582 webpage = self._download_webpage(
583 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
584 return self.extract_ytcfg(video_id, webpage) or {}
585
586 @staticmethod
587 def _build_api_continuation_query(continuation, ctp=None):
588 query = {
589 'continuation': continuation
590 }
591 # TODO: Inconsistency with clickTrackingParams.
592 # Currently we have a fixed ctp contained within context (from ytcfg)
593 # and a ctp in root query for continuation.
594 if ctp:
595 query['clickTracking'] = {'clickTrackingParams': ctp}
596 return query
597
598 @classmethod
599 def _extract_next_continuation_data(cls, renderer):
600 next_continuation = try_get(
601 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
602 lambda x: x['continuation']['reloadContinuationData']), dict)
603 if not next_continuation:
604 return
605 continuation = next_continuation.get('continuation')
606 if not continuation:
607 return
608 ctp = next_continuation.get('clickTrackingParams')
609 return cls._build_api_continuation_query(continuation, ctp)
610
611 @classmethod
612 def _extract_continuation_ep_data(cls, continuation_ep: dict):
613 if isinstance(continuation_ep, dict):
614 continuation = try_get(
615 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
616 if not continuation:
617 return
618 ctp = continuation_ep.get('clickTrackingParams')
619 return cls._build_api_continuation_query(continuation, ctp)
620
621 @classmethod
622 def _extract_continuation(cls, renderer):
623 next_continuation = cls._extract_next_continuation_data(renderer)
624 if next_continuation:
625 return next_continuation
626
627 contents = []
628 for key in ('contents', 'items'):
629 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
630
631 for content in contents:
632 if not isinstance(content, dict):
633 continue
634 continuation_ep = try_get(
635 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
636 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
637 dict)
638 continuation = cls._extract_continuation_ep_data(continuation_ep)
639 if continuation:
640 return continuation
641
642 @classmethod
643 def _extract_alerts(cls, data):
644 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
645 if not isinstance(alert_dict, dict):
646 continue
647 for alert in alert_dict.values():
648 alert_type = alert.get('type')
649 if not alert_type:
650 continue
651 message = cls._get_text(alert, 'text')
652 if message:
653 yield alert_type, message
654
655 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
656 errors = []
657 warnings = []
658 for alert_type, alert_message in alerts:
659 if alert_type.lower() == 'error' and fatal:
660 errors.append([alert_type, alert_message])
661 else:
662 warnings.append([alert_type, alert_message])
663
664 for alert_type, alert_message in (warnings + errors[:-1]):
665 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
666 if errors:
667 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
668
669 def _extract_and_report_alerts(self, data, *args, **kwargs):
670 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
671
672 def _extract_badges(self, renderer: dict):
673 badges = set()
674 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
675 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
676 if label:
677 badges.add(label.lower())
678 return badges
679
680 @staticmethod
681 def _get_text(data, *path_list, max_runs=None):
682 for path in path_list or [None]:
683 if path is None:
684 obj = [data]
685 else:
686 obj = traverse_obj(data, path, default=[])
687 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
688 obj = [obj]
689 for item in obj:
690 text = try_get(item, lambda x: x['simpleText'], compat_str)
691 if text:
692 return text
693 runs = try_get(item, lambda x: x['runs'], list) or []
694 if not runs and isinstance(item, list):
695 runs = item
696
697 runs = runs[:min(len(runs), max_runs or len(runs))]
698 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
699 if text:
700 return text
701
702 def _get_count(self, data, *path_list):
703 count_text = self._get_text(data, *path_list) or ''
704 count = parse_count(count_text)
705 if count is None:
706 count = str_to_int(
707 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
708 return count
709
710 @staticmethod
711 def _extract_thumbnails(data, *path_list):
712 """
713 Extract thumbnails from thumbnails dict
714 @param path_list: path list to level that contains 'thumbnails' key
715 """
716 thumbnails = []
717 for path in path_list or [()]:
718 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
719 thumbnail_url = url_or_none(thumbnail.get('url'))
720 if not thumbnail_url:
721 continue
722 # Sometimes youtube gives a wrong thumbnail URL. See:
723 # https://github.com/yt-dlp/yt-dlp/issues/233
724 # https://github.com/ytdl-org/youtube-dl/issues/28023
725 if 'maxresdefault' in thumbnail_url:
726 thumbnail_url = thumbnail_url.split('?')[0]
727 thumbnails.append({
728 'url': thumbnail_url,
729 'height': int_or_none(thumbnail.get('height')),
730 'width': int_or_none(thumbnail.get('width')),
731 })
732 return thumbnails
733
734 @staticmethod
735 def extract_relative_time(relative_time_text):
736 """
737 Extracts a relative time from string and converts to dt object
738 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
739 """
740 mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
741 if mobj:
742 start = mobj.group('start')
743 if start:
744 return datetime_from_str(start)
745 try:
746 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
747 except ValueError:
748 return None
749
750 def _extract_time_text(self, renderer, *path_list):
751 """@returns (timestamp, time_text)"""
752 text = self._get_text(renderer, *path_list) or ''
753 dt = self.extract_relative_time(text)
754 timestamp = None
755 if isinstance(dt, datetime.datetime):
756 timestamp = calendar.timegm(dt.timetuple())
757
758 if timestamp is None:
759 timestamp = (
760 unified_timestamp(text) or unified_timestamp(
761 self._search_regex(
762 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
763 text.lower(), 'time text', default=None)))
764
765 if text and timestamp is None:
766 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
767 return timestamp, text
768
769 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
770 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
771 default_client='web'):
772 response = None
773 last_error = None
774 count = -1
775 retries = self.get_param('extractor_retries', 3)
776 if check_get_keys is None:
777 check_get_keys = []
778 while count < retries:
779 count += 1
780 if last_error:
781 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
782 try:
783 response = self._call_api(
784 ep=ep, fatal=True, headers=headers,
785 video_id=item_id, query=query,
786 context=self._extract_context(ytcfg, default_client),
787 api_key=self._extract_api_key(ytcfg, default_client),
788 api_hostname=api_hostname, default_client=default_client,
789 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
790 except ExtractorError as e:
791 if isinstance(e.cause, network_exceptions):
792 if isinstance(e.cause, compat_HTTPError):
793 first_bytes = e.cause.read(512)
794 if not is_html(first_bytes):
795 yt_error = try_get(
796 self._parse_json(
797 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
798 lambda x: x['error']['message'], compat_str)
799 if yt_error:
800 self._report_alerts([('ERROR', yt_error)], fatal=False)
801 # Downloading page may result in intermittent 5xx HTTP error
802 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
803 # We also want to catch all other network exceptions since errors in later pages can be troublesome
804 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
805 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
806 last_error = error_to_compat_str(e.cause or e.msg)
807 if count < retries:
808 continue
809 if fatal:
810 raise
811 else:
812 self.report_warning(error_to_compat_str(e))
813 return
814
815 else:
816 try:
817 self._extract_and_report_alerts(response, only_once=True)
818 except ExtractorError as e:
819 # YouTube servers may return errors we want to retry on in a 200 OK response
820 # See: https://github.com/yt-dlp/yt-dlp/issues/839
821 if 'unknown error' in e.msg.lower():
822 last_error = e.msg
823 continue
824 if fatal:
825 raise
826 self.report_warning(error_to_compat_str(e))
827 return
828 if not check_get_keys or dict_get(response, check_get_keys):
829 break
830 # Youtube sometimes sends incomplete data
831 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
832 last_error = 'Incomplete data received'
833 if count >= retries:
834 if fatal:
835 raise ExtractorError(last_error)
836 else:
837 self.report_warning(last_error)
838 return
839 return response
840
841 @staticmethod
842 def is_music_url(url):
843 return re.match(r'https?://music\.youtube\.com/', url) is not None
844
845 def _extract_video(self, renderer):
846 video_id = renderer.get('videoId')
847 title = self._get_text(renderer, 'title')
848 description = self._get_text(renderer, 'descriptionSnippet')
849 duration = parse_duration(self._get_text(
850 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
851 if duration is None:
852 duration = parse_duration(self._search_regex(
853 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
854 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
855 video_id, default=None, group='duration'))
856
857 view_count = self._get_count(renderer, 'viewCountText')
858
859 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
860 channel_id = traverse_obj(
861 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
862 expected_type=str, get_all=False)
863 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
864 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
865 overlay_style = traverse_obj(
866 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
867 get_all=False, expected_type=str)
868 badges = self._extract_badges(renderer)
869 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
870 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
871 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
872 expected_type=str)) or ''
873 url = f'https://www.youtube.com/watch?v={video_id}'
874 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
875 url = f'https://www.youtube.com/shorts/{video_id}'
876
877 return {
878 '_type': 'url',
879 'ie_key': YoutubeIE.ie_key(),
880 'id': video_id,
881 'url': url,
882 'title': title,
883 'description': description,
884 'duration': duration,
885 'view_count': view_count,
886 'uploader': uploader,
887 'channel_id': channel_id,
888 'thumbnails': thumbnails,
889 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
890 if self._configuration_arg('approximate_date', ie_key='youtubetab')
891 else None),
892 'live_status': ('is_upcoming' if scheduled_timestamp is not None
893 else 'was_live' if 'streamed' in time_text.lower()
894 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
895 else None),
896 'release_timestamp': scheduled_timestamp,
897 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
898 }
899
900
901 class YoutubeIE(YoutubeBaseInfoExtractor):
902 IE_DESC = 'YouTube'
903 _VALID_URL = r"""(?x)^
904 (
905 (?:https?://|//) # http(s):// or protocol-independent URL
906 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
907 (?:www\.)?deturl\.com/www\.youtube\.com|
908 (?:www\.)?pwnyoutube\.com|
909 (?:www\.)?hooktube\.com|
910 (?:www\.)?yourepeat\.com|
911 tube\.majestyc\.net|
912 %(invidious)s|
913 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
914 (?:.*?\#/)? # handle anchor (#/) redirect urls
915 (?: # the various things that can precede the ID:
916 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
917 |(?: # or the v= param in all its forms
918 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
919 (?:\?|\#!?) # the params delimiter ? or # or #!
920 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
921 v=
922 )
923 ))
924 |(?:
925 youtu\.be| # just youtu.be/xxxx
926 vid\.plus| # or vid.plus/xxxx
927 zwearz\.com/watch| # or zwearz.com/watch/xxxx
928 %(invidious)s
929 )/
930 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
931 )
932 )? # all until now is optional -> you can pass the naked ID
933 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
934 (?(1).+)? # if we found the ID, everything can follow
935 (?:\#|$)""" % {
936 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
937 }
938 _PLAYER_INFO_RE = (
939 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
940 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
941 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
942 )
943 _formats = {
944 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
945 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
946 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
947 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
948 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
949 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
950 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
951 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
952 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
953 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
954 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
955 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
956 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
957 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
958 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
959 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
960 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
961 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
962
963
964 # 3D videos
965 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
966 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
967 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
968 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
969 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
970 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
971 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
972
973 # Apple HTTP Live Streaming
974 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
975 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
976 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
977 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
978 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
979 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
980 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
981 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
982
983 # DASH mp4 video
984 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
985 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
986 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
987 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
988 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
989 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
990 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
991 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
992 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
993 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
994 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
995 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
996
997 # Dash mp4 audio
998 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
999 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1000 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1001 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1002 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1003 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1004 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1005
1006 # Dash webm
1007 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1008 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1009 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1010 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1011 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1012 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1013 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1014 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1015 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1016 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1017 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1018 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1019 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1020 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1021 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1022 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1023 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1024 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1025 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1026 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1027 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1028 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1029
1030 # Dash webm audio
1031 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1032 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1033
1034 # Dash webm audio with opus inside
1035 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1036 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1037 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1038
1039 # RTMP (unnamed)
1040 '_rtmp': {'protocol': 'rtmp'},
1041
1042 # av01 video only formats sometimes served with "unknown" codecs
1043 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1044 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1045 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1046 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1047 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1048 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1049 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1050 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1051 }
1052 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1053
1054 _GEO_BYPASS = False
1055
1056 IE_NAME = 'youtube'
1057 _TESTS = [
1058 {
1059 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1060 'info_dict': {
1061 'id': 'BaW_jenozKc',
1062 'ext': 'mp4',
1063 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1064 'uploader': 'Philipp Hagemeister',
1065 'uploader_id': 'phihag',
1066 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1067 'channel': 'Philipp Hagemeister',
1068 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1069 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1070 'upload_date': '20121002',
1071 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1072 'categories': ['Science & Technology'],
1073 'tags': ['youtube-dl'],
1074 'duration': 10,
1075 'view_count': int,
1076 'like_count': int,
1077 'availability': 'public',
1078 'playable_in_embed': True,
1079 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1080 'live_status': 'not_live',
1081 'age_limit': 0,
1082 'start_time': 1,
1083 'end_time': 9,
1084 'channel_follower_count': int
1085 }
1086 },
1087 {
1088 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1089 'note': 'Embed-only video (#1746)',
1090 'info_dict': {
1091 'id': 'yZIXLfi8CZQ',
1092 'ext': 'mp4',
1093 'upload_date': '20120608',
1094 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1095 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1096 'uploader': 'SET India',
1097 'uploader_id': 'setindia',
1098 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1099 'age_limit': 18,
1100 },
1101 'skip': 'Private video',
1102 },
1103 {
1104 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1105 'note': 'Use the first video ID in the URL',
1106 'info_dict': {
1107 'id': 'BaW_jenozKc',
1108 'ext': 'mp4',
1109 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1110 'uploader': 'Philipp Hagemeister',
1111 'uploader_id': 'phihag',
1112 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1113 'channel': 'Philipp Hagemeister',
1114 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1115 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1116 'upload_date': '20121002',
1117 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1118 'categories': ['Science & Technology'],
1119 'tags': ['youtube-dl'],
1120 'duration': 10,
1121 'view_count': int,
1122 'like_count': int,
1123 'availability': 'public',
1124 'playable_in_embed': True,
1125 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1126 'live_status': 'not_live',
1127 'age_limit': 0,
1128 'channel_follower_count': int
1129 },
1130 'params': {
1131 'skip_download': True,
1132 },
1133 },
1134 {
1135 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1136 'note': '256k DASH audio (format 141) via DASH manifest',
1137 'info_dict': {
1138 'id': 'a9LDPn-MO4I',
1139 'ext': 'm4a',
1140 'upload_date': '20121002',
1141 'uploader_id': '8KVIDEO',
1142 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1143 'description': '',
1144 'uploader': '8KVIDEO',
1145 'title': 'UHDTV TEST 8K VIDEO.mp4'
1146 },
1147 'params': {
1148 'youtube_include_dash_manifest': True,
1149 'format': '141',
1150 },
1151 'skip': 'format 141 not served anymore',
1152 },
1153 # DASH manifest with encrypted signature
1154 {
1155 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1156 'info_dict': {
1157 'id': 'IB3lcPjvWLA',
1158 'ext': 'm4a',
1159 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1160 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1161 'duration': 244,
1162 'uploader': 'AfrojackVEVO',
1163 'uploader_id': 'AfrojackVEVO',
1164 'upload_date': '20131011',
1165 'abr': 129.495,
1166 'like_count': int,
1167 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1168 'playable_in_embed': True,
1169 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1170 'view_count': int,
1171 'track': 'The Spark',
1172 'live_status': 'not_live',
1173 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1174 'channel': 'Afrojack',
1175 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1176 'tags': 'count:19',
1177 'availability': 'public',
1178 'categories': ['Music'],
1179 'age_limit': 0,
1180 'alt_title': 'The Spark',
1181 'channel_follower_count': int
1182 },
1183 'params': {
1184 'youtube_include_dash_manifest': True,
1185 'format': '141/bestaudio[ext=m4a]',
1186 },
1187 },
1188 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1189 {
1190 'note': 'Embed allowed age-gate video',
1191 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1192 'info_dict': {
1193 'id': 'HtVdAasjOgU',
1194 'ext': 'mp4',
1195 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1196 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1197 'duration': 142,
1198 'uploader': 'The Witcher',
1199 'uploader_id': 'WitcherGame',
1200 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1201 'upload_date': '20140605',
1202 'age_limit': 18,
1203 'categories': ['Gaming'],
1204 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1205 'availability': 'needs_auth',
1206 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1207 'like_count': int,
1208 'channel': 'The Witcher',
1209 'live_status': 'not_live',
1210 'tags': 'count:17',
1211 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1212 'playable_in_embed': True,
1213 'view_count': int,
1214 'channel_follower_count': int
1215 },
1216 },
1217 {
1218 'note': 'Age-gate video with embed allowed in public site',
1219 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1220 'info_dict': {
1221 'id': 'HsUATh_Nc2U',
1222 'ext': 'mp4',
1223 'title': 'Godzilla 2 (Official Video)',
1224 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1225 'upload_date': '20200408',
1226 'uploader_id': 'FlyingKitty900',
1227 'uploader': 'FlyingKitty',
1228 'age_limit': 18,
1229 'availability': 'needs_auth',
1230 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1231 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1232 'channel': 'FlyingKitty',
1233 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1234 'view_count': int,
1235 'categories': ['Entertainment'],
1236 'live_status': 'not_live',
1237 'tags': ['Flyingkitty', 'godzilla 2'],
1238 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1239 'like_count': int,
1240 'duration': 177,
1241 'playable_in_embed': True,
1242 'channel_follower_count': int
1243 },
1244 },
1245 {
1246 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1247 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1248 'info_dict': {
1249 'id': 'Tq92D6wQ1mg',
1250 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1251 'ext': 'mp4',
1252 'upload_date': '20191228',
1253 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1254 'uploader': 'Projekt Melody',
1255 'description': 'md5:17eccca93a786d51bc67646756894066',
1256 'age_limit': 18,
1257 'like_count': int,
1258 'availability': 'needs_auth',
1259 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1260 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1261 'view_count': int,
1262 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1263 'channel': 'Projekt Melody',
1264 'live_status': 'not_live',
1265 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1266 'playable_in_embed': True,
1267 'categories': ['Entertainment'],
1268 'duration': 106,
1269 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1270 'channel_follower_count': int
1271 },
1272 },
1273 {
1274 'note': 'Non-Agegated non-embeddable video',
1275 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1276 'info_dict': {
1277 'id': 'MeJVWBSsPAY',
1278 'ext': 'mp4',
1279 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1280 'uploader': 'Herr Lurik',
1281 'uploader_id': 'st3in234',
1282 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1283 'upload_date': '20130730',
1284 'track': 'Such mich find mich',
1285 'age_limit': 0,
1286 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1287 'like_count': int,
1288 'playable_in_embed': False,
1289 'creator': 'OOMPH!',
1290 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1291 'view_count': int,
1292 'alt_title': 'Such mich find mich',
1293 'duration': 210,
1294 'channel': 'Herr Lurik',
1295 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1296 'categories': ['Music'],
1297 'availability': 'public',
1298 'uploader_url': 'http://www.youtube.com/user/st3in234',
1299 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1300 'live_status': 'not_live',
1301 'artist': 'OOMPH!',
1302 'channel_follower_count': int
1303 },
1304 },
1305 {
1306 'note': 'Non-bypassable age-gated video',
1307 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1308 'only_matching': True,
1309 },
1310 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1311 # YouTube Red ad is not captured for creator
1312 {
1313 'url': '__2ABJjxzNo',
1314 'info_dict': {
1315 'id': '__2ABJjxzNo',
1316 'ext': 'mp4',
1317 'duration': 266,
1318 'upload_date': '20100430',
1319 'uploader_id': 'deadmau5',
1320 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1321 'creator': 'deadmau5',
1322 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1323 'uploader': 'deadmau5',
1324 'title': 'Deadmau5 - Some Chords (HD)',
1325 'alt_title': 'Some Chords',
1326 'availability': 'public',
1327 'tags': 'count:14',
1328 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1329 'view_count': int,
1330 'live_status': 'not_live',
1331 'channel': 'deadmau5',
1332 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1333 'like_count': int,
1334 'track': 'Some Chords',
1335 'artist': 'deadmau5',
1336 'playable_in_embed': True,
1337 'age_limit': 0,
1338 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1339 'categories': ['Music'],
1340 'album': 'Some Chords',
1341 'channel_follower_count': int
1342 },
1343 'expected_warnings': [
1344 'DASH manifest missing',
1345 ]
1346 },
1347 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1348 {
1349 'url': 'lqQg6PlCWgI',
1350 'info_dict': {
1351 'id': 'lqQg6PlCWgI',
1352 'ext': 'mp4',
1353 'duration': 6085,
1354 'upload_date': '20150827',
1355 'uploader_id': 'olympic',
1356 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1357 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1358 'uploader': 'Olympics',
1359 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1360 'like_count': int,
1361 'release_timestamp': 1343767800,
1362 'playable_in_embed': True,
1363 'categories': ['Sports'],
1364 'release_date': '20120731',
1365 'channel': 'Olympics',
1366 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1367 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1368 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1369 'age_limit': 0,
1370 'availability': 'public',
1371 'live_status': 'was_live',
1372 'view_count': int,
1373 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1374 'channel_follower_count': int
1375 },
1376 'params': {
1377 'skip_download': 'requires avconv',
1378 }
1379 },
1380 # Non-square pixels
1381 {
1382 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1383 'info_dict': {
1384 'id': '_b-2C3KPAM0',
1385 'ext': 'mp4',
1386 'stretched_ratio': 16 / 9.,
1387 'duration': 85,
1388 'upload_date': '20110310',
1389 'uploader_id': 'AllenMeow',
1390 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1391 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
1392 'uploader': '孫ᄋᄅ',
1393 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1394 'playable_in_embed': True,
1395 'channel': '孫ᄋᄅ',
1396 'age_limit': 0,
1397 'tags': 'count:11',
1398 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1399 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1400 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1401 'view_count': int,
1402 'categories': ['People & Blogs'],
1403 'like_count': int,
1404 'live_status': 'not_live',
1405 'availability': 'unlisted',
1406 'channel_follower_count': int
1407 },
1408 },
1409 # url_encoded_fmt_stream_map is empty string
1410 {
1411 'url': 'qEJwOuvDf7I',
1412 'info_dict': {
1413 'id': 'qEJwOuvDf7I',
1414 'ext': 'webm',
1415 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1416 'description': '',
1417 'upload_date': '20150404',
1418 'uploader_id': 'spbelect',
1419 'uploader': 'Наблюдатели Петербурга',
1420 },
1421 'params': {
1422 'skip_download': 'requires avconv',
1423 },
1424 'skip': 'This live event has ended.',
1425 },
1426 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1427 {
1428 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1429 'info_dict': {
1430 'id': 'FIl7x6_3R5Y',
1431 'ext': 'webm',
1432 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1433 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1434 'duration': 220,
1435 'upload_date': '20150625',
1436 'uploader_id': 'dorappi2000',
1437 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1438 'uploader': 'dorappi2000',
1439 'formats': 'mincount:31',
1440 },
1441 'skip': 'not actual anymore',
1442 },
1443 # DASH manifest with segment_list
1444 {
1445 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1446 'md5': '8ce563a1d667b599d21064e982ab9e31',
1447 'info_dict': {
1448 'id': 'CsmdDsKjzN8',
1449 'ext': 'mp4',
1450 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1451 'uploader': 'Airtek',
1452 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1453 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1454 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1455 },
1456 'params': {
1457 'youtube_include_dash_manifest': True,
1458 'format': '135', # bestvideo
1459 },
1460 'skip': 'This live event has ended.',
1461 },
1462 {
1463 # Multifeed videos (multiple cameras), URL is for Main Camera
1464 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1465 'info_dict': {
1466 'id': 'jvGDaLqkpTg',
1467 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1468 'description': 'md5:e03b909557865076822aa169218d6a5d',
1469 },
1470 'playlist': [{
1471 'info_dict': {
1472 'id': 'jvGDaLqkpTg',
1473 'ext': 'mp4',
1474 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1475 'description': 'md5:e03b909557865076822aa169218d6a5d',
1476 'duration': 10643,
1477 'upload_date': '20161111',
1478 'uploader': 'Team PGP',
1479 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1480 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1481 },
1482 }, {
1483 'info_dict': {
1484 'id': '3AKt1R1aDnw',
1485 'ext': 'mp4',
1486 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1487 'description': 'md5:e03b909557865076822aa169218d6a5d',
1488 'duration': 10991,
1489 'upload_date': '20161111',
1490 'uploader': 'Team PGP',
1491 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1492 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1493 },
1494 }, {
1495 'info_dict': {
1496 'id': 'RtAMM00gpVc',
1497 'ext': 'mp4',
1498 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1499 'description': 'md5:e03b909557865076822aa169218d6a5d',
1500 'duration': 10995,
1501 'upload_date': '20161111',
1502 'uploader': 'Team PGP',
1503 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1504 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1505 },
1506 }, {
1507 'info_dict': {
1508 'id': '6N2fdlP3C5U',
1509 'ext': 'mp4',
1510 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1511 'description': 'md5:e03b909557865076822aa169218d6a5d',
1512 'duration': 10990,
1513 'upload_date': '20161111',
1514 'uploader': 'Team PGP',
1515 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1516 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1517 },
1518 }],
1519 'params': {
1520 'skip_download': True,
1521 },
1522 'skip': 'Not multifeed anymore',
1523 },
1524 {
1525 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1526 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1527 'info_dict': {
1528 'id': 'gVfLd0zydlo',
1529 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1530 },
1531 'playlist_count': 2,
1532 'skip': 'Not multifeed anymore',
1533 },
1534 {
1535 'url': 'https://vid.plus/FlRa-iH7PGw',
1536 'only_matching': True,
1537 },
1538 {
1539 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1540 'only_matching': True,
1541 },
1542 {
1543 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1544 # Also tests cut-off URL expansion in video description (see
1545 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1546 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1547 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1548 'info_dict': {
1549 'id': 'lsguqyKfVQg',
1550 'ext': 'mp4',
1551 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1552 'alt_title': 'Dark Walk',
1553 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1554 'duration': 133,
1555 'upload_date': '20151119',
1556 'uploader_id': 'IronSoulElf',
1557 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1558 'uploader': 'IronSoulElf',
1559 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1560 'track': 'Dark Walk',
1561 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1562 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1563 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1564 'categories': ['Film & Animation'],
1565 'view_count': int,
1566 'live_status': 'not_live',
1567 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1568 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1569 'tags': 'count:13',
1570 'availability': 'public',
1571 'channel': 'IronSoulElf',
1572 'playable_in_embed': True,
1573 'like_count': int,
1574 'age_limit': 0,
1575 'channel_follower_count': int
1576 },
1577 'params': {
1578 'skip_download': True,
1579 },
1580 },
1581 {
1582 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1583 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1584 'only_matching': True,
1585 },
1586 {
1587 # Video with yt:stretch=17:0
1588 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1589 'info_dict': {
1590 'id': 'Q39EVAstoRM',
1591 'ext': 'mp4',
1592 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1593 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1594 'upload_date': '20151107',
1595 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1596 'uploader': 'CH GAMER DROID',
1597 },
1598 'params': {
1599 'skip_download': True,
1600 },
1601 'skip': 'This video does not exist.',
1602 },
1603 {
1604 # Video with incomplete 'yt:stretch=16:'
1605 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1606 'only_matching': True,
1607 },
1608 {
1609 # Video licensed under Creative Commons
1610 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1611 'info_dict': {
1612 'id': 'M4gD1WSo5mA',
1613 'ext': 'mp4',
1614 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1615 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1616 'duration': 721,
1617 'upload_date': '20150128',
1618 'uploader_id': 'BerkmanCenter',
1619 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1620 'uploader': 'The Berkman Klein Center for Internet & Society',
1621 'license': 'Creative Commons Attribution license (reuse allowed)',
1622 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1623 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1624 'like_count': int,
1625 'age_limit': 0,
1626 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1627 'channel': 'The Berkman Klein Center for Internet & Society',
1628 'availability': 'public',
1629 'view_count': int,
1630 'categories': ['Education'],
1631 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1632 'live_status': 'not_live',
1633 'playable_in_embed': True,
1634 'channel_follower_count': int
1635 },
1636 'params': {
1637 'skip_download': True,
1638 },
1639 },
1640 {
1641 # Channel-like uploader_url
1642 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1643 'info_dict': {
1644 'id': 'eQcmzGIKrzg',
1645 'ext': 'mp4',
1646 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1647 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1648 'duration': 4060,
1649 'upload_date': '20151120',
1650 'uploader': 'Bernie Sanders',
1651 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1652 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1653 'license': 'Creative Commons Attribution license (reuse allowed)',
1654 'playable_in_embed': True,
1655 'tags': 'count:12',
1656 'like_count': int,
1657 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1658 'age_limit': 0,
1659 'availability': 'public',
1660 'categories': ['News & Politics'],
1661 'channel': 'Bernie Sanders',
1662 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1663 'view_count': int,
1664 'live_status': 'not_live',
1665 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1666 'channel_follower_count': int
1667 },
1668 'params': {
1669 'skip_download': True,
1670 },
1671 },
1672 {
1673 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1674 'only_matching': True,
1675 },
1676 {
1677 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1678 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1679 'only_matching': True,
1680 },
1681 {
1682 # Rental video preview
1683 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1684 'info_dict': {
1685 'id': 'uGpuVWrhIzE',
1686 'ext': 'mp4',
1687 'title': 'Piku - Trailer',
1688 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1689 'upload_date': '20150811',
1690 'uploader': 'FlixMatrix',
1691 'uploader_id': 'FlixMatrixKaravan',
1692 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1693 'license': 'Standard YouTube License',
1694 },
1695 'params': {
1696 'skip_download': True,
1697 },
1698 'skip': 'This video is not available.',
1699 },
1700 {
1701 # YouTube Red video with episode data
1702 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1703 'info_dict': {
1704 'id': 'iqKdEhx-dD4',
1705 'ext': 'mp4',
1706 'title': 'Isolation - Mind Field (Ep 1)',
1707 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1708 'duration': 2085,
1709 'upload_date': '20170118',
1710 'uploader': 'Vsauce',
1711 'uploader_id': 'Vsauce',
1712 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1713 'series': 'Mind Field',
1714 'season_number': 1,
1715 'episode_number': 1,
1716 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1717 'tags': 'count:12',
1718 'view_count': int,
1719 'availability': 'public',
1720 'age_limit': 0,
1721 'channel': 'Vsauce',
1722 'episode': 'Episode 1',
1723 'categories': ['Entertainment'],
1724 'season': 'Season 1',
1725 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1726 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1727 'like_count': int,
1728 'playable_in_embed': True,
1729 'live_status': 'not_live',
1730 'channel_follower_count': int
1731 },
1732 'params': {
1733 'skip_download': True,
1734 },
1735 'expected_warnings': [
1736 'Skipping DASH manifest',
1737 ],
1738 },
1739 {
1740 # The following content has been identified by the YouTube community
1741 # as inappropriate or offensive to some audiences.
1742 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1743 'info_dict': {
1744 'id': '6SJNVb0GnPI',
1745 'ext': 'mp4',
1746 'title': 'Race Differences in Intelligence',
1747 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1748 'duration': 965,
1749 'upload_date': '20140124',
1750 'uploader': 'New Century Foundation',
1751 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1752 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1753 },
1754 'params': {
1755 'skip_download': True,
1756 },
1757 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1758 },
1759 {
1760 # itag 212
1761 'url': '1t24XAntNCY',
1762 'only_matching': True,
1763 },
1764 {
1765 # geo restricted to JP
1766 'url': 'sJL6WA-aGkQ',
1767 'only_matching': True,
1768 },
1769 {
1770 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1771 'only_matching': True,
1772 },
1773 {
1774 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1775 'only_matching': True,
1776 },
1777 {
1778 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1779 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1780 'only_matching': True,
1781 },
1782 {
1783 # DRM protected
1784 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1785 'only_matching': True,
1786 },
1787 {
1788 # Video with unsupported adaptive stream type formats
1789 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1790 'info_dict': {
1791 'id': 'Z4Vy8R84T1U',
1792 'ext': 'mp4',
1793 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1794 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1795 'duration': 433,
1796 'upload_date': '20130923',
1797 'uploader': 'Amelia Putri Harwita',
1798 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1799 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1800 'formats': 'maxcount:10',
1801 },
1802 'params': {
1803 'skip_download': True,
1804 'youtube_include_dash_manifest': False,
1805 },
1806 'skip': 'not actual anymore',
1807 },
1808 {
1809 # Youtube Music Auto-generated description
1810 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1811 'info_dict': {
1812 'id': 'MgNrAu2pzNs',
1813 'ext': 'mp4',
1814 'title': 'Voyeur Girl',
1815 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1816 'upload_date': '20190312',
1817 'uploader': 'Stephen - Topic',
1818 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1819 'artist': 'Stephen',
1820 'track': 'Voyeur Girl',
1821 'album': 'it\'s too much love to know my dear',
1822 'release_date': '20190313',
1823 'release_year': 2019,
1824 'alt_title': 'Voyeur Girl',
1825 'view_count': int,
1826 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1827 'playable_in_embed': True,
1828 'like_count': int,
1829 'categories': ['Music'],
1830 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1831 'channel': 'Stephen',
1832 'availability': 'public',
1833 'creator': 'Stephen',
1834 'duration': 169,
1835 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1836 'age_limit': 0,
1837 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1838 'tags': 'count:11',
1839 'live_status': 'not_live',
1840 'channel_follower_count': int
1841 },
1842 'params': {
1843 'skip_download': True,
1844 },
1845 },
1846 {
1847 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1848 'only_matching': True,
1849 },
1850 {
1851 # invalid -> valid video id redirection
1852 'url': 'DJztXj2GPfl',
1853 'info_dict': {
1854 'id': 'DJztXj2GPfk',
1855 'ext': 'mp4',
1856 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1857 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1858 'upload_date': '20090125',
1859 'uploader': 'Prochorowka',
1860 'uploader_id': 'Prochorowka',
1861 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1862 'artist': 'Panjabi MC',
1863 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1864 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1865 },
1866 'params': {
1867 'skip_download': True,
1868 },
1869 'skip': 'Video unavailable',
1870 },
1871 {
1872 # empty description results in an empty string
1873 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1874 'info_dict': {
1875 'id': 'x41yOUIvK2k',
1876 'ext': 'mp4',
1877 'title': 'IMG 3456',
1878 'description': '',
1879 'upload_date': '20170613',
1880 'uploader_id': 'ElevageOrVert',
1881 'uploader': 'ElevageOrVert',
1882 'view_count': int,
1883 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1884 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1885 'like_count': int,
1886 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1887 'tags': [],
1888 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1889 'availability': 'public',
1890 'age_limit': 0,
1891 'categories': ['Pets & Animals'],
1892 'duration': 7,
1893 'playable_in_embed': True,
1894 'live_status': 'not_live',
1895 'channel': 'ElevageOrVert',
1896 'channel_follower_count': int
1897 },
1898 'params': {
1899 'skip_download': True,
1900 },
1901 },
1902 {
1903 # with '};' inside yt initial data (see [1])
1904 # see [2] for an example with '};' inside ytInitialPlayerResponse
1905 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1906 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1907 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1908 'info_dict': {
1909 'id': 'CHqg6qOn4no',
1910 'ext': 'mp4',
1911 'title': 'Part 77 Sort a list of simple types in c#',
1912 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1913 'upload_date': '20130831',
1914 'uploader_id': 'kudvenkat',
1915 'uploader': 'kudvenkat',
1916 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1917 'like_count': int,
1918 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1919 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1920 'live_status': 'not_live',
1921 'categories': ['Education'],
1922 'availability': 'public',
1923 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1924 'tags': 'count:12',
1925 'playable_in_embed': True,
1926 'age_limit': 0,
1927 'view_count': int,
1928 'duration': 522,
1929 'channel': 'kudvenkat',
1930 'channel_follower_count': int
1931 },
1932 'params': {
1933 'skip_download': True,
1934 },
1935 },
1936 {
1937 # another example of '};' in ytInitialData
1938 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1939 'only_matching': True,
1940 },
1941 {
1942 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1943 'only_matching': True,
1944 },
1945 {
1946 # https://github.com/ytdl-org/youtube-dl/pull/28094
1947 'url': 'OtqTfy26tG0',
1948 'info_dict': {
1949 'id': 'OtqTfy26tG0',
1950 'ext': 'mp4',
1951 'title': 'Burn Out',
1952 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1953 'upload_date': '20141120',
1954 'uploader': 'The Cinematic Orchestra - Topic',
1955 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1956 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1957 'artist': 'The Cinematic Orchestra',
1958 'track': 'Burn Out',
1959 'album': 'Every Day',
1960 'like_count': int,
1961 'live_status': 'not_live',
1962 'alt_title': 'Burn Out',
1963 'duration': 614,
1964 'age_limit': 0,
1965 'view_count': int,
1966 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1967 'creator': 'The Cinematic Orchestra',
1968 'channel': 'The Cinematic Orchestra',
1969 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1970 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1971 'availability': 'public',
1972 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1973 'categories': ['Music'],
1974 'playable_in_embed': True,
1975 'channel_follower_count': int
1976 },
1977 'params': {
1978 'skip_download': True,
1979 },
1980 },
1981 {
1982 # controversial video, only works with bpctr when authenticated with cookies
1983 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1984 'only_matching': True,
1985 },
1986 {
1987 # controversial video, requires bpctr/contentCheckOk
1988 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1989 'info_dict': {
1990 'id': 'SZJvDhaSDnc',
1991 'ext': 'mp4',
1992 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1993 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1994 'uploader': 'CBS Mornings',
1995 'uploader_id': 'CBSThisMorning',
1996 'upload_date': '20140716',
1997 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
1998 'duration': 170,
1999 'categories': ['News & Politics'],
2000 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2001 'view_count': int,
2002 'channel': 'CBS Mornings',
2003 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2004 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2005 'age_limit': 18,
2006 'availability': 'needs_auth',
2007 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2008 'like_count': int,
2009 'live_status': 'not_live',
2010 'playable_in_embed': True,
2011 'channel_follower_count': int
2012 }
2013 },
2014 {
2015 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2016 'url': 'cBvYw8_A0vQ',
2017 'info_dict': {
2018 'id': 'cBvYw8_A0vQ',
2019 'ext': 'mp4',
2020 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2021 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2022 'upload_date': '20201120',
2023 'uploader': 'Walk around Japan',
2024 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2026 'duration': 1456,
2027 'categories': ['Travel & Events'],
2028 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2029 'view_count': int,
2030 'channel': 'Walk around Japan',
2031 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2032 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2033 'age_limit': 0,
2034 'availability': 'public',
2035 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2036 'live_status': 'not_live',
2037 'playable_in_embed': True,
2038 'channel_follower_count': int
2039 },
2040 'params': {
2041 'skip_download': True,
2042 },
2043 }, {
2044 # Has multiple audio streams
2045 'url': 'WaOKSUlf4TM',
2046 'only_matching': True
2047 }, {
2048 # Requires Premium: has format 141 when requested using YTM url
2049 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2050 'only_matching': True
2051 }, {
2052 # multiple subtitles with same lang_code
2053 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2054 'only_matching': True,
2055 }, {
2056 # Force use android client fallback
2057 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2058 'info_dict': {
2059 'id': 'YOelRv7fMxY',
2060 'title': 'DIGGING A SECRET TUNNEL Part 1',
2061 'ext': '3gp',
2062 'upload_date': '20210624',
2063 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2064 'uploader': 'colinfurze',
2065 'uploader_id': 'colinfurze',
2066 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2067 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2068 'duration': 596,
2069 'categories': ['Entertainment'],
2070 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2071 'view_count': int,
2072 'channel': 'colinfurze',
2073 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2074 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2075 'age_limit': 0,
2076 'availability': 'public',
2077 'like_count': int,
2078 'live_status': 'not_live',
2079 'playable_in_embed': True,
2080 'channel_follower_count': int
2081 },
2082 'params': {
2083 'format': '17', # 3gp format available on android
2084 'extractor_args': {'youtube': {'player_client': ['android']}},
2085 },
2086 },
2087 {
2088 # Skip download of additional client configs (remix client config in this case)
2089 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2090 'only_matching': True,
2091 'params': {
2092 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2093 },
2094 }, {
2095 # shorts
2096 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2097 'only_matching': True,
2098 }, {
2099 'note': 'Storyboards',
2100 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2101 'info_dict': {
2102 'id': '5KLPxDtMqe8',
2103 'ext': 'mhtml',
2104 'format_id': 'sb0',
2105 'title': 'Your Brain is Plastic',
2106 'uploader_id': 'scishow',
2107 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2108 'upload_date': '20140324',
2109 'uploader': 'SciShow',
2110 'like_count': int,
2111 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2112 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2113 'view_count': int,
2114 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2115 'playable_in_embed': True,
2116 'tags': 'count:12',
2117 'uploader_url': 'http://www.youtube.com/user/scishow',
2118 'availability': 'public',
2119 'channel': 'SciShow',
2120 'live_status': 'not_live',
2121 'duration': 248,
2122 'categories': ['Education'],
2123 'age_limit': 0,
2124 'channel_follower_count': int
2125 }, 'params': {'format': 'mhtml', 'skip_download': True}
2126 }, {
2127 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2128 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2129 'info_dict': {
2130 'id': '2NUZ8W2llS4',
2131 'ext': 'mp4',
2132 'title': 'The NP that test your phone performance 🙂',
2133 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2134 'uploader': 'Leon Nguyen',
2135 'uploader_id': 'VNSXIII',
2136 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2137 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2138 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2139 'duration': 21,
2140 'view_count': int,
2141 'age_limit': 0,
2142 'categories': ['Gaming'],
2143 'tags': 'count:23',
2144 'playable_in_embed': True,
2145 'live_status': 'not_live',
2146 'upload_date': '20220103',
2147 'like_count': int,
2148 'availability': 'public',
2149 'channel': 'Leon Nguyen',
2150 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2151 'channel_follower_count': int
2152 }
2153 }, {
2154 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2155 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2156 'info_dict': {
2157 'id': 'mzZzzBU6lrM',
2158 'ext': 'mp4',
2159 'title': 'I Met GeorgeNotFound In Real Life...',
2160 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2161 'uploader': 'Quackity',
2162 'uploader_id': 'QuackityHQ',
2163 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2164 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2165 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2166 'duration': 955,
2167 'view_count': int,
2168 'age_limit': 0,
2169 'categories': ['Entertainment'],
2170 'tags': 'count:26',
2171 'playable_in_embed': True,
2172 'live_status': 'not_live',
2173 'release_timestamp': 1641172509,
2174 'release_date': '20220103',
2175 'upload_date': '20220103',
2176 'like_count': int,
2177 'availability': 'public',
2178 'channel': 'Quackity',
2179 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2180 'channel_follower_count': int
2181 }
2182 },
2183 { # continuous livestream. Microformat upload date should be preferred.
2184 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2185 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2186 'info_dict': {
2187 'id': 'kgx4WGK0oNU',
2188 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2189 'ext': 'mp4',
2190 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2191 'availability': 'public',
2192 'age_limit': 0,
2193 'release_timestamp': 1637975704,
2194 'upload_date': '20210619',
2195 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2196 'live_status': 'is_live',
2197 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2198 'uploader': '阿鲍Abao',
2199 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2200 'channel': 'Abao in Tokyo',
2201 'channel_follower_count': int,
2202 'release_date': '20211127',
2203 'tags': 'count:39',
2204 'categories': ['People & Blogs'],
2205 'like_count': int,
2206 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2207 'view_count': int,
2208 'playable_in_embed': True,
2209 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2210 },
2211 'params': {'skip_download': True}
2212 }, {
2213 # Story. Requires specific player params to work.
2214 # Note: stories get removed after some period of time
2215 'url': 'https://www.youtube.com/watch?v=yN3x1t3sieA',
2216 'info_dict': {
2217 'id': 'yN3x1t3sieA',
2218 'ext': 'mp4',
2219 'uploader': 'Linus Tech Tips',
2220 'duration': 13,
2221 'channel': 'Linus Tech Tips',
2222 'playable_in_embed': True,
2223 'tags': [],
2224 'age_limit': 0,
2225 'uploader_url': 'http://www.youtube.com/user/LinusTechTips',
2226 'upload_date': '20220402',
2227 'thumbnail': 'https://i.ytimg.com/vi_webp/yN3x1t3sieA/maxresdefault.webp',
2228 'title': 'Story',
2229 'live_status': 'not_live',
2230 'uploader_id': 'LinusTechTips',
2231 'view_count': int,
2232 'description': '',
2233 'channel_id': 'UCXuqSBlHAE6Xw-yeJA0Tunw',
2234 'categories': ['Science & Technology'],
2235 'channel_url': 'https://www.youtube.com/channel/UCXuqSBlHAE6Xw-yeJA0Tunw',
2236 'availability': 'unlisted',
2237 }
2238 }
2239 ]
2240
2241 @classmethod
2242 def suitable(cls, url):
2243 from ..utils import parse_qs
2244
2245 qs = parse_qs(url)
2246 if qs.get('list', [None])[0]:
2247 return False
2248 return super().suitable(url)
2249
2250 def __init__(self, *args, **kwargs):
2251 super().__init__(*args, **kwargs)
2252 self._code_cache = {}
2253 self._player_cache = {}
2254
2255 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2256 lock = threading.Lock()
2257
2258 is_live = True
2259 start_time = time.time()
2260 formats = [f for f in formats if f.get('is_from_start')]
2261
2262 def refetch_manifest(format_id, delay):
2263 nonlocal formats, start_time, is_live
2264 if time.time() <= start_time + delay:
2265 return
2266
2267 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2268 video_details = traverse_obj(
2269 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2270 microformats = traverse_obj(
2271 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2272 expected_type=dict, default=[])
2273 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2274 start_time = time.time()
2275
2276 def mpd_feed(format_id, delay):
2277 """
2278 @returns (manifest_url, manifest_stream_number, is_live) or None
2279 """
2280 with lock:
2281 refetch_manifest(format_id, delay)
2282
2283 f = next((f for f in formats if f['format_id'] == format_id), None)
2284 if not f:
2285 if not is_live:
2286 self.to_screen(f'{video_id}: Video is no longer live')
2287 else:
2288 self.report_warning(
2289 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2290 return None
2291 return f['manifest_url'], f['manifest_stream_number'], is_live
2292
2293 for f in formats:
2294 f['is_live'] = True
2295 f['protocol'] = 'http_dash_segments_generator'
2296 f['fragments'] = functools.partial(
2297 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2298
2299 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2300 FETCH_SPAN, MAX_DURATION = 5, 432000
2301
2302 mpd_url, stream_number, is_live = None, None, True
2303
2304 begin_index = 0
2305 download_start_time = ctx.get('start') or time.time()
2306
2307 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2308 if lack_early_segments:
2309 self.report_warning(bug_reports_message(
2310 'Starting download from the last 120 hours of the live stream since '
2311 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2312 lack_early_segments = True
2313
2314 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2315 fragments, fragment_base_url = None, None
2316
2317 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2318 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2319 # Obtain from MPD's maximum seq value
2320 old_mpd_url = mpd_url
2321 last_error = ctx.pop('last_error', None)
2322 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2323 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2324 or (mpd_url, stream_number, False))
2325 if not refresh_sequence:
2326 if expire_fast and not is_live:
2327 return False, last_seq
2328 elif old_mpd_url == mpd_url:
2329 return True, last_seq
2330 try:
2331 fmts, _ = self._extract_mpd_formats_and_subtitles(
2332 mpd_url, None, note=False, errnote=False, fatal=False)
2333 except ExtractorError:
2334 fmts = None
2335 if not fmts:
2336 no_fragment_score += 2
2337 return False, last_seq
2338 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2339 fragments = fmt_info['fragments']
2340 fragment_base_url = fmt_info['fragment_base_url']
2341 assert fragment_base_url
2342
2343 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2344 return True, _last_seq
2345
2346 while is_live:
2347 fetch_time = time.time()
2348 if no_fragment_score > 30:
2349 return
2350 if last_segment_url:
2351 # Obtain from "X-Head-Seqnum" header value from each segment
2352 try:
2353 urlh = self._request_webpage(
2354 last_segment_url, None, note=False, errnote=False, fatal=False)
2355 except ExtractorError:
2356 urlh = None
2357 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2358 if last_seq is None:
2359 no_fragment_score += 2
2360 last_segment_url = None
2361 continue
2362 else:
2363 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2364 no_fragment_score += 2
2365 if not should_continue:
2366 continue
2367
2368 if known_idx > last_seq:
2369 last_segment_url = None
2370 continue
2371
2372 last_seq += 1
2373
2374 if begin_index < 0 and known_idx < 0:
2375 # skip from the start when it's negative value
2376 known_idx = last_seq + begin_index
2377 if lack_early_segments:
2378 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2379 try:
2380 for idx in range(known_idx, last_seq):
2381 # do not update sequence here or you'll get skipped some part of it
2382 should_continue, _ = _extract_sequence_from_mpd(False, False)
2383 if not should_continue:
2384 known_idx = idx - 1
2385 raise ExtractorError('breaking out of outer loop')
2386 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2387 yield {
2388 'url': last_segment_url,
2389 }
2390 if known_idx == last_seq:
2391 no_fragment_score += 5
2392 else:
2393 no_fragment_score = 0
2394 known_idx = last_seq
2395 except ExtractorError:
2396 continue
2397
2398 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2399
2400 def _extract_player_url(self, *ytcfgs, webpage=None):
2401 player_url = traverse_obj(
2402 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2403 get_all=False, expected_type=compat_str)
2404 if not player_url:
2405 return
2406 return urljoin('https://www.youtube.com', player_url)
2407
2408 def _download_player_url(self, video_id, fatal=False):
2409 res = self._download_webpage(
2410 'https://www.youtube.com/iframe_api',
2411 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2412 if res:
2413 player_version = self._search_regex(
2414 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2415 if player_version:
2416 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2417
2418 def _signature_cache_id(self, example_sig):
2419 """ Return a string representation of a signature """
2420 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2421
2422 @classmethod
2423 def _extract_player_info(cls, player_url):
2424 for player_re in cls._PLAYER_INFO_RE:
2425 id_m = re.search(player_re, player_url)
2426 if id_m:
2427 break
2428 else:
2429 raise ExtractorError('Cannot identify player %r' % player_url)
2430 return id_m.group('id')
2431
2432 def _load_player(self, video_id, player_url, fatal=True):
2433 player_id = self._extract_player_info(player_url)
2434 if player_id not in self._code_cache:
2435 code = self._download_webpage(
2436 player_url, video_id, fatal=fatal,
2437 note='Downloading player ' + player_id,
2438 errnote='Download of %s failed' % player_url)
2439 if code:
2440 self._code_cache[player_id] = code
2441 return self._code_cache.get(player_id)
2442
2443 def _extract_signature_function(self, video_id, player_url, example_sig):
2444 player_id = self._extract_player_info(player_url)
2445
2446 # Read from filesystem cache
2447 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2448 assert os.path.basename(func_id) == func_id
2449
2450 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2451 if cache_spec is not None:
2452 return lambda s: ''.join(s[i] for i in cache_spec)
2453
2454 code = self._load_player(video_id, player_url)
2455 if code:
2456 res = self._parse_sig_js(code)
2457
2458 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2459 cache_res = res(test_string)
2460 cache_spec = [ord(c) for c in cache_res]
2461
2462 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2463 return res
2464
2465 def _print_sig_code(self, func, example_sig):
2466 if not self.get_param('youtube_print_sig_code'):
2467 return
2468
2469 def gen_sig_code(idxs):
2470 def _genslice(start, end, step):
2471 starts = '' if start == 0 else str(start)
2472 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2473 steps = '' if step == 1 else (':%d' % step)
2474 return f's[{starts}{ends}{steps}]'
2475
2476 step = None
2477 # Quelch pyflakes warnings - start will be set when step is set
2478 start = '(Never used)'
2479 for i, prev in zip(idxs[1:], idxs[:-1]):
2480 if step is not None:
2481 if i - prev == step:
2482 continue
2483 yield _genslice(start, prev, step)
2484 step = None
2485 continue
2486 if i - prev in [-1, 1]:
2487 step = i - prev
2488 start = prev
2489 continue
2490 else:
2491 yield 's[%d]' % prev
2492 if step is None:
2493 yield 's[%d]' % i
2494 else:
2495 yield _genslice(start, i, step)
2496
2497 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2498 cache_res = func(test_string)
2499 cache_spec = [ord(c) for c in cache_res]
2500 expr_code = ' + '.join(gen_sig_code(cache_spec))
2501 signature_id_tuple = '(%s)' % (
2502 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2503 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2504 ' return %s\n') % (signature_id_tuple, expr_code)
2505 self.to_screen('Extracted signature function:\n' + code)
2506
2507 def _parse_sig_js(self, jscode):
2508 funcname = self._search_regex(
2509 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2510 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2511 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2512 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2513 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)',
2514 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2515 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2516 # Obsolete patterns
2517 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2518 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2519 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2520 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2521 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2522 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2523 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2524 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2525 jscode, 'Initial JS player signature function name', group='sig')
2526
2527 jsi = JSInterpreter(jscode)
2528 initial_function = jsi.extract_function(funcname)
2529 return lambda s: initial_function([s])
2530
2531 def _decrypt_signature(self, s, video_id, player_url):
2532 """Turn the encrypted s field into a working signature"""
2533 try:
2534 player_id = (player_url, self._signature_cache_id(s))
2535 if player_id not in self._player_cache:
2536 func = self._extract_signature_function(video_id, player_url, s)
2537 self._player_cache[player_id] = func
2538 func = self._player_cache[player_id]
2539 self._print_sig_code(func, s)
2540 return func(s)
2541 except Exception as e:
2542 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2543
2544 def _decrypt_nsig(self, s, video_id, player_url):
2545 """Turn the encrypted n field into a working signature"""
2546 if player_url is None:
2547 raise ExtractorError('Cannot decrypt nsig without player_url')
2548 player_url = urljoin('https://www.youtube.com', player_url)
2549
2550 sig_id = ('nsig_value', s)
2551 if sig_id in self._player_cache:
2552 return self._player_cache[sig_id]
2553
2554 try:
2555 player_id = ('nsig', player_url)
2556 if player_id not in self._player_cache:
2557 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2558 func = self._player_cache[player_id]
2559 self._player_cache[sig_id] = func(s)
2560 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2561 return self._player_cache[sig_id]
2562 except Exception as e:
2563 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2564
2565 def _extract_n_function_name(self, jscode):
2566 nfunc, idx = self._search_regex(
2567 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2568 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2569 if not idx:
2570 return nfunc
2571 return json.loads(js_to_json(self._search_regex(
2572 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2573 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2574
2575 def _extract_n_function(self, video_id, player_url):
2576 player_id = self._extract_player_info(player_url)
2577 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2578
2579 if func_code:
2580 jsi = JSInterpreter(func_code)
2581 else:
2582 jscode = self._load_player(video_id, player_url)
2583 funcname = self._extract_n_function_name(jscode)
2584 jsi = JSInterpreter(jscode)
2585 func_code = jsi.extract_function_code(funcname)
2586 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2587
2588 if self.get_param('youtube_print_sig_code'):
2589 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2590
2591 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2592
2593 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2594 """
2595 Extract signatureTimestamp (sts)
2596 Required to tell API what sig/player version is in use.
2597 """
2598 sts = None
2599 if isinstance(ytcfg, dict):
2600 sts = int_or_none(ytcfg.get('STS'))
2601
2602 if not sts:
2603 # Attempt to extract from player
2604 if player_url is None:
2605 error_msg = 'Cannot extract signature timestamp without player_url.'
2606 if fatal:
2607 raise ExtractorError(error_msg)
2608 self.report_warning(error_msg)
2609 return
2610 code = self._load_player(video_id, player_url, fatal=fatal)
2611 if code:
2612 sts = int_or_none(self._search_regex(
2613 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2614 'JS player signature timestamp', group='sts', fatal=fatal))
2615 return sts
2616
2617 def _mark_watched(self, video_id, player_responses):
2618 playback_url = get_first(
2619 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2620 expected_type=url_or_none)
2621 if not playback_url:
2622 self.report_warning('Unable to mark watched')
2623 return
2624 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2625 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2626
2627 # cpn generation algorithm is reverse engineered from base.js.
2628 # In fact it works even with dummy cpn.
2629 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2630 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2631
2632 qs.update({
2633 'ver': ['2'],
2634 'cpn': [cpn],
2635 })
2636 playback_url = compat_urlparse.urlunparse(
2637 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2638
2639 self._download_webpage(
2640 playback_url, video_id, 'Marking watched',
2641 'Unable to mark watched', fatal=False)
2642
2643 @staticmethod
2644 def _extract_urls(webpage):
2645 # Embedded YouTube player
2646 entries = [
2647 unescapeHTML(mobj.group('url'))
2648 for mobj in re.finditer(r'''(?x)
2649 (?:
2650 <iframe[^>]+?src=|
2651 data-video-url=|
2652 <embed[^>]+?src=|
2653 embedSWF\(?:\s*|
2654 <object[^>]+data=|
2655 new\s+SWFObject\(
2656 )
2657 (["\'])
2658 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2659 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2660 \1''', webpage)]
2661
2662 # lazyYT YouTube embed
2663 entries.extend(list(map(
2664 unescapeHTML,
2665 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2666
2667 # Wordpress "YouTube Video Importer" plugin
2668 matches = re.findall(r'''(?x)<div[^>]+
2669 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2670 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2671 entries.extend(m[-1] for m in matches)
2672
2673 return entries
2674
2675 @staticmethod
2676 def _extract_url(webpage):
2677 urls = YoutubeIE._extract_urls(webpage)
2678 return urls[0] if urls else None
2679
2680 @classmethod
2681 def extract_id(cls, url):
2682 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2683 if mobj is None:
2684 raise ExtractorError('Invalid URL: %s' % url)
2685 return mobj.group('id')
2686
2687 def _extract_chapters_from_json(self, data, duration):
2688 chapter_list = traverse_obj(
2689 data, (
2690 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2691 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2692 ), expected_type=list)
2693
2694 return self._extract_chapters(
2695 chapter_list,
2696 chapter_time=lambda chapter: float_or_none(
2697 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2698 chapter_title=lambda chapter: traverse_obj(
2699 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2700 duration=duration)
2701
2702 def _extract_chapters_from_engagement_panel(self, data, duration):
2703 content_list = traverse_obj(
2704 data,
2705 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2706 expected_type=list, default=[])
2707 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2708 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2709
2710 return next((
2711 filter(None, (
2712 self._extract_chapters(
2713 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2714 chapter_time, chapter_title, duration)
2715 for contents in content_list
2716 ))), [])
2717
2718 @staticmethod
2719 def _extract_chapters_from_description(description, duration):
2720 chapters = [{'start_time': 0}]
2721 for timestamp, title in re.findall(
2722 r'(?m)^((?:\d+:)?\d{1,2}:\d{2})\b\W*\s(.+?)\s*$', description or ''):
2723 start = parse_duration(timestamp)
2724 if start and title and chapters[-1]['start_time'] < start < duration:
2725 chapters[-1]['end_time'] = start
2726 chapters.append({
2727 'start_time': start,
2728 'title': title,
2729 })
2730 chapters[-1]['end_time'] = duration
2731 return chapters[1:]
2732
2733 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2734 chapters = []
2735 last_chapter = {'start_time': 0}
2736 for idx, chapter in enumerate(chapter_list or []):
2737 title = chapter_title(chapter)
2738 start_time = chapter_time(chapter)
2739 if start_time is None:
2740 continue
2741 last_chapter['end_time'] = start_time
2742 if start_time < last_chapter['start_time']:
2743 if idx == 1:
2744 chapters.pop()
2745 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2746 else:
2747 self.report_warning(f'Invalid start time for chapter "{title}"')
2748 continue
2749 last_chapter = {'start_time': start_time, 'title': title}
2750 chapters.append(last_chapter)
2751 last_chapter['end_time'] = duration
2752 return chapters
2753
2754 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2755 return self._parse_json(self._search_regex(
2756 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
2757 regex), webpage, name, default='{}'), video_id, fatal=False)
2758
2759 def _extract_comment(self, comment_renderer, parent=None):
2760 comment_id = comment_renderer.get('commentId')
2761 if not comment_id:
2762 return
2763
2764 text = self._get_text(comment_renderer, 'contentText')
2765
2766 # note: timestamp is an estimate calculated from the current time and time_text
2767 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2768 author = self._get_text(comment_renderer, 'authorText')
2769 author_id = try_get(comment_renderer,
2770 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2771
2772 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2773 lambda x: x['likeCount']), compat_str)) or 0
2774 author_thumbnail = try_get(comment_renderer,
2775 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2776
2777 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2778 is_favorited = 'creatorHeart' in (try_get(
2779 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2780 return {
2781 'id': comment_id,
2782 'text': text,
2783 'timestamp': timestamp,
2784 'time_text': time_text,
2785 'like_count': votes,
2786 'is_favorited': is_favorited,
2787 'author': author,
2788 'author_id': author_id,
2789 'author_thumbnail': author_thumbnail,
2790 'author_is_uploader': author_is_uploader,
2791 'parent': parent or 'root'
2792 }
2793
2794 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2795
2796 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2797
2798 def extract_header(contents):
2799 _continuation = None
2800 for content in contents:
2801 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2802 expected_comment_count = self._get_count(
2803 comments_header_renderer, 'countText', 'commentsCount')
2804
2805 if expected_comment_count:
2806 tracker['est_total'] = expected_comment_count
2807 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2808 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2809
2810 sort_menu_item = try_get(
2811 comments_header_renderer,
2812 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2813 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2814
2815 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2816 if not _continuation:
2817 continue
2818
2819 sort_text = str_or_none(sort_menu_item.get('title'))
2820 if not sort_text:
2821 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2822 self.to_screen('Sorting comments by %s' % sort_text.lower())
2823 break
2824 return _continuation
2825
2826 def extract_thread(contents):
2827 if not parent:
2828 tracker['current_page_thread'] = 0
2829 for content in contents:
2830 if not parent and tracker['total_parent_comments'] >= max_parents:
2831 yield
2832 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2833 comment_renderer = get_first(
2834 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2835 expected_type=dict, default={})
2836
2837 comment = self._extract_comment(comment_renderer, parent)
2838 if not comment:
2839 continue
2840
2841 tracker['running_total'] += 1
2842 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2843 yield comment
2844
2845 # Attempt to get the replies
2846 comment_replies_renderer = try_get(
2847 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2848
2849 if comment_replies_renderer:
2850 tracker['current_page_thread'] += 1
2851 comment_entries_iter = self._comment_entries(
2852 comment_replies_renderer, ytcfg, video_id,
2853 parent=comment.get('id'), tracker=tracker)
2854 yield from itertools.islice(comment_entries_iter, min(
2855 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2856
2857 # Keeps track of counts across recursive calls
2858 if not tracker:
2859 tracker = dict(
2860 running_total=0,
2861 est_total=0,
2862 current_page_thread=0,
2863 total_parent_comments=0,
2864 total_reply_comments=0)
2865
2866 # TODO: Deprecated
2867 # YouTube comments have a max depth of 2
2868 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2869 if max_depth:
2870 self._downloader.deprecation_warning(
2871 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2872 if max_depth == 1 and parent:
2873 return
2874
2875 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2876 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2877
2878 continuation = self._extract_continuation(root_continuation_data)
2879
2880 response = None
2881 is_forced_continuation = False
2882 is_first_continuation = parent is None
2883 if is_first_continuation and not continuation:
2884 # Sometimes you can get comments by generating the continuation yourself,
2885 # even if YouTube initially reports them being disabled - e.g. stories comments.
2886 # Note: if the comment section is actually disabled, YouTube may return a response with
2887 # required check_get_keys missing. So we will disable that check initially in this case.
2888 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2889 is_forced_continuation = True
2890
2891 for page_num in itertools.count(0):
2892 if not continuation:
2893 break
2894 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2895 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2896 if page_num == 0:
2897 if is_first_continuation:
2898 note_prefix = 'Downloading comment section API JSON'
2899 else:
2900 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2901 tracker['current_page_thread'], comment_prog_str)
2902 else:
2903 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2904 ' ' if parent else '', ' replies' if parent else '',
2905 page_num, comment_prog_str)
2906
2907 response = self._extract_response(
2908 item_id=None, query=continuation,
2909 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2910 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2911 is_forced_continuation = False
2912 continuation_contents = traverse_obj(
2913 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2914
2915 continuation = None
2916 for continuation_section in continuation_contents:
2917 continuation_items = traverse_obj(
2918 continuation_section,
2919 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2920 get_all=False, expected_type=list) or []
2921 if is_first_continuation:
2922 continuation = extract_header(continuation_items)
2923 is_first_continuation = False
2924 if continuation:
2925 break
2926 continue
2927
2928 for entry in extract_thread(continuation_items):
2929 if not entry:
2930 return
2931 yield entry
2932 continuation = self._extract_continuation({'contents': continuation_items})
2933 if continuation:
2934 break
2935
2936 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2937 if message and not parent and tracker['running_total'] == 0:
2938 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2939
2940 @staticmethod
2941 def _generate_comment_continuation(video_id):
2942 """
2943 Generates initial comment section continuation token from given video id
2944 """
2945 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2946 return base64.b64encode(token.encode()).decode()
2947
2948 def _get_comments(self, ytcfg, video_id, contents, webpage):
2949 """Entry for comment extraction"""
2950 def _real_comment_extract(contents):
2951 renderer = next((
2952 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2953 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2954 yield from self._comment_entries(renderer, ytcfg, video_id)
2955
2956 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2957 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2958
2959 @staticmethod
2960 def _get_checkok_params():
2961 return {'contentCheckOk': True, 'racyCheckOk': True}
2962
2963 @classmethod
2964 def _generate_player_context(cls, sts=None):
2965 context = {
2966 'html5Preference': 'HTML5_PREF_WANTS',
2967 }
2968 if sts is not None:
2969 context['signatureTimestamp'] = sts
2970 return {
2971 'playbackContext': {
2972 'contentPlaybackContext': context
2973 },
2974 **cls._get_checkok_params()
2975 }
2976
2977 @staticmethod
2978 def _is_agegated(player_response):
2979 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2980 return True
2981
2982 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2983 AGE_GATE_REASONS = (
2984 'confirm your age', 'age-restricted', 'inappropriate', # reason
2985 'age_verification_required', 'age_check_required', # status
2986 )
2987 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2988
2989 @staticmethod
2990 def _is_unplayable(player_response):
2991 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2992
2993 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2994
2995 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2996 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2997 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2998 headers = self.generate_api_headers(
2999 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3000
3001 yt_query = {
3002 'videoId': video_id,
3003 'params': '8AEB' # enable stories
3004 }
3005 yt_query.update(self._generate_player_context(sts))
3006 return self._extract_response(
3007 item_id=video_id, ep='player', query=yt_query,
3008 ytcfg=player_ytcfg, headers=headers, fatal=True,
3009 default_client=client,
3010 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3011 ) or None
3012
3013 def _get_requested_clients(self, url, smuggled_data):
3014 requested_clients = []
3015 default = ['android', 'web']
3016 allowed_clients = sorted(
3017 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3018 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3019 for client in self._configuration_arg('player_client'):
3020 if client in allowed_clients:
3021 requested_clients.append(client)
3022 elif client == 'default':
3023 requested_clients.extend(default)
3024 elif client == 'all':
3025 requested_clients.extend(allowed_clients)
3026 else:
3027 self.report_warning(f'Skipping unsupported client {client}')
3028 if not requested_clients:
3029 requested_clients = default
3030
3031 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3032 requested_clients.extend(
3033 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3034
3035 return orderedSet(requested_clients)
3036
3037 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3038 initial_pr = None
3039 if webpage:
3040 initial_pr = self._extract_yt_initial_variable(
3041 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
3042 video_id, 'initial player response')
3043
3044 all_clients = set(clients)
3045 clients = clients[::-1]
3046 prs = []
3047
3048 def append_client(*client_names):
3049 """ Append the first client name that exists but not already used """
3050 for client_name in client_names:
3051 actual_client = _split_innertube_client(client_name)[0]
3052 if actual_client in INNERTUBE_CLIENTS:
3053 if actual_client not in all_clients:
3054 clients.append(client_name)
3055 all_clients.add(actual_client)
3056 return
3057
3058 # Android player_response does not have microFormats which are needed for
3059 # extraction of some data. So we return the initial_pr with formats
3060 # stripped out even if not requested by the user
3061 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3062 if initial_pr:
3063 pr = dict(initial_pr)
3064 pr['streamingData'] = None
3065 prs.append(pr)
3066
3067 last_error = None
3068 tried_iframe_fallback = False
3069 player_url = None
3070 while clients:
3071 client, base_client, variant = _split_innertube_client(clients.pop())
3072 player_ytcfg = master_ytcfg if client == 'web' else {}
3073 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3074 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3075
3076 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3077 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3078 if 'js' in self._configuration_arg('player_skip'):
3079 require_js_player = False
3080 player_url = None
3081
3082 if not player_url and not tried_iframe_fallback and require_js_player:
3083 player_url = self._download_player_url(video_id)
3084 tried_iframe_fallback = True
3085
3086 try:
3087 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3088 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3089 except ExtractorError as e:
3090 if last_error:
3091 self.report_warning(last_error)
3092 last_error = e
3093 continue
3094
3095 if pr:
3096 prs.append(pr)
3097
3098 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3099 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3100 append_client(f'{base_client}_creator')
3101 elif self._is_agegated(pr):
3102 if variant == 'tv_embedded':
3103 append_client(f'{base_client}_embedded')
3104 elif not variant:
3105 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3106
3107 if last_error:
3108 if not len(prs):
3109 raise last_error
3110 self.report_warning(last_error)
3111 return prs, player_url
3112
3113 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3114 itags, stream_ids = {}, []
3115 itag_qualities, res_qualities = {}, {}
3116 q = qualities([
3117 # Normally tiny is the smallest video-only formats. But
3118 # audio-only formats with unknown quality may get tagged as tiny
3119 'tiny',
3120 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3121 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3122 ])
3123 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3124
3125 for fmt in streaming_formats:
3126 if fmt.get('targetDurationSec'):
3127 continue
3128
3129 itag = str_or_none(fmt.get('itag'))
3130 audio_track = fmt.get('audioTrack') or {}
3131 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3132 if stream_id in stream_ids:
3133 continue
3134
3135 quality = fmt.get('quality')
3136 height = int_or_none(fmt.get('height'))
3137 if quality == 'tiny' or not quality:
3138 quality = fmt.get('audioQuality', '').lower() or quality
3139 # The 3gp format (17) in android client has a quality of "small",
3140 # but is actually worse than other formats
3141 if itag == '17':
3142 quality = 'tiny'
3143 if quality:
3144 if itag:
3145 itag_qualities[itag] = quality
3146 if height:
3147 res_qualities[height] = quality
3148 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3149 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3150 # number of fragment that would subsequently requested with (`&sq=N`)
3151 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3152 continue
3153
3154 fmt_url = fmt.get('url')
3155 if not fmt_url:
3156 sc = compat_parse_qs(fmt.get('signatureCipher'))
3157 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3158 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3159 if not all((sc, fmt_url, player_url, encrypted_sig)):
3160 continue
3161 try:
3162 fmt_url += '&%s=%s' % (
3163 traverse_obj(sc, ('sp', -1)) or 'signature',
3164 self._decrypt_signature(encrypted_sig, video_id, player_url)
3165 )
3166 except ExtractorError as e:
3167 self.report_warning('Signature extraction failed: Some formats may be missing', only_once=True)
3168 self.write_debug(e, only_once=True)
3169 continue
3170
3171 query = parse_qs(fmt_url)
3172 throttled = False
3173 if query.get('n'):
3174 try:
3175 fmt_url = update_url_query(fmt_url, {
3176 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3177 except ExtractorError as e:
3178 self.report_warning(
3179 'nsig extraction failed: You may experience throttling for some formats\n'
3180 f'n = {query["n"][0]} ; player = {player_url}', only_once=True)
3181 self.write_debug(e, only_once=True)
3182 throttled = True
3183
3184 if itag:
3185 itags[itag] = 'https'
3186 stream_ids.append(stream_id)
3187
3188 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3189 language_preference = (
3190 10 if audio_track.get('audioIsDefault') and 10
3191 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3192 else -1)
3193 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3194 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3195 # Make sure to avoid false positives with small duration differences.
3196 # Eg: __2ABJjxzNo, ySuUZEjARPY
3197 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3198 if is_damaged:
3199 self.report_warning(
3200 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3201 dct = {
3202 'asr': int_or_none(fmt.get('audioSampleRate')),
3203 'filesize': int_or_none(fmt.get('contentLength')),
3204 'format_id': itag,
3205 'format_note': join_nonempty(
3206 '%s%s' % (audio_track.get('displayName') or '',
3207 ' (default)' if language_preference > 0 else ''),
3208 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3209 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3210 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3211 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3212 'fps': int_or_none(fmt.get('fps')) or None,
3213 'height': height,
3214 'quality': q(quality),
3215 'has_drm': bool(fmt.get('drmFamilies')),
3216 'tbr': tbr,
3217 'url': fmt_url,
3218 'width': int_or_none(fmt.get('width')),
3219 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3220 'desc' if language_preference < -1 else ''),
3221 'language_preference': language_preference,
3222 # Strictly de-prioritize damaged and 3gp formats
3223 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3224 }
3225 mime_mobj = re.match(
3226 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3227 if mime_mobj:
3228 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3229 dct.update(parse_codecs(mime_mobj.group(2)))
3230 no_audio = dct.get('acodec') == 'none'
3231 no_video = dct.get('vcodec') == 'none'
3232 if no_audio:
3233 dct['vbr'] = tbr
3234 if no_video:
3235 dct['abr'] = tbr
3236 if no_audio or no_video:
3237 dct['downloader_options'] = {
3238 # Youtube throttles chunks >~10M
3239 'http_chunk_size': 10485760,
3240 }
3241 if dct.get('ext'):
3242 dct['container'] = dct['ext'] + '_dash'
3243 yield dct
3244
3245 live_from_start = is_live and self.get_param('live_from_start')
3246 skip_manifests = self._configuration_arg('skip')
3247 if not self.get_param('youtube_include_hls_manifest', True):
3248 skip_manifests.append('hls')
3249 if not self.get_param('youtube_include_dash_manifest', True):
3250 skip_manifests.append('dash')
3251 get_dash = 'dash' not in skip_manifests and (
3252 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3253 get_hls = not live_from_start and 'hls' not in skip_manifests
3254
3255 def process_manifest_format(f, proto, itag):
3256 if itag in itags:
3257 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3258 return False
3259 itag = f'{itag}-{proto}'
3260 if itag:
3261 f['format_id'] = itag
3262 itags[itag] = proto
3263
3264 f['quality'] = next((
3265 q(qdict[val])
3266 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3267 if val in qdict), -1)
3268 return True
3269
3270 for sd in streaming_data:
3271 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3272 if hls_manifest_url:
3273 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3274 if process_manifest_format(f, 'hls', self._search_regex(
3275 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3276 yield f
3277
3278 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3279 if dash_manifest_url:
3280 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3281 if process_manifest_format(f, 'dash', f['format_id']):
3282 f['filesize'] = int_or_none(self._search_regex(
3283 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3284 if live_from_start:
3285 f['is_from_start'] = True
3286
3287 yield f
3288
3289 def _extract_storyboard(self, player_responses, duration):
3290 spec = get_first(
3291 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3292 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3293 if not base_url:
3294 return
3295 L = len(spec) - 1
3296 for i, args in enumerate(spec):
3297 args = args.split('#')
3298 counts = list(map(int_or_none, args[:5]))
3299 if len(args) != 8 or not all(counts):
3300 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3301 continue
3302 width, height, frame_count, cols, rows = counts
3303 N, sigh = args[6:]
3304
3305 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3306 fragment_count = frame_count / (cols * rows)
3307 fragment_duration = duration / fragment_count
3308 yield {
3309 'format_id': f'sb{i}',
3310 'format_note': 'storyboard',
3311 'ext': 'mhtml',
3312 'protocol': 'mhtml',
3313 'acodec': 'none',
3314 'vcodec': 'none',
3315 'url': url,
3316 'width': width,
3317 'height': height,
3318 'fragments': [{
3319 'url': url.replace('$M', str(j)),
3320 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3321 } for j in range(math.ceil(fragment_count))],
3322 }
3323
3324 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3325 webpage = None
3326 if 'webpage' not in self._configuration_arg('player_skip'):
3327 webpage = self._download_webpage(
3328 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3329
3330 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3331
3332 player_responses, player_url = self._extract_player_responses(
3333 self._get_requested_clients(url, smuggled_data),
3334 video_id, webpage, master_ytcfg)
3335
3336 return webpage, master_ytcfg, player_responses, player_url
3337
3338 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3339 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3340 is_live = get_first(video_details, 'isLive')
3341 if is_live is None:
3342 is_live = get_first(live_broadcast_details, 'isLiveNow')
3343
3344 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3345 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3346
3347 return live_broadcast_details, is_live, streaming_data, formats
3348
3349 def _real_extract(self, url):
3350 url, smuggled_data = unsmuggle_url(url, {})
3351 video_id = self._match_id(url)
3352
3353 base_url = self.http_scheme() + '//www.youtube.com/'
3354 webpage_url = base_url + 'watch?v=' + video_id
3355
3356 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3357
3358 playability_statuses = traverse_obj(
3359 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3360
3361 trailer_video_id = get_first(
3362 playability_statuses,
3363 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3364 expected_type=str)
3365 if trailer_video_id:
3366 return self.url_result(
3367 trailer_video_id, self.ie_key(), trailer_video_id)
3368
3369 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3370 if webpage else (lambda x: None))
3371
3372 video_details = traverse_obj(
3373 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3374 microformats = traverse_obj(
3375 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3376 expected_type=dict, default=[])
3377 video_title = (
3378 get_first(video_details, 'title')
3379 or self._get_text(microformats, (..., 'title'))
3380 or search_meta(['og:title', 'twitter:title', 'title']))
3381 video_description = get_first(video_details, 'shortDescription')
3382
3383 multifeed_metadata_list = get_first(
3384 player_responses,
3385 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3386 expected_type=str)
3387 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3388 if self.get_param('noplaylist'):
3389 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3390 else:
3391 entries = []
3392 feed_ids = []
3393 for feed in multifeed_metadata_list.split(','):
3394 # Unquote should take place before split on comma (,) since textual
3395 # fields may contain comma as well (see
3396 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3397 feed_data = compat_parse_qs(
3398 compat_urllib_parse_unquote_plus(feed))
3399
3400 def feed_entry(name):
3401 return try_get(
3402 feed_data, lambda x: x[name][0], compat_str)
3403
3404 feed_id = feed_entry('id')
3405 if not feed_id:
3406 continue
3407 feed_title = feed_entry('title')
3408 title = video_title
3409 if feed_title:
3410 title += ' (%s)' % feed_title
3411 entries.append({
3412 '_type': 'url_transparent',
3413 'ie_key': 'Youtube',
3414 'url': smuggle_url(
3415 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3416 {'force_singlefeed': True}),
3417 'title': title,
3418 })
3419 feed_ids.append(feed_id)
3420 self.to_screen(
3421 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3422 % (', '.join(feed_ids), video_id))
3423 return self.playlist_result(
3424 entries, video_id, video_title, video_description)
3425
3426 duration = int_or_none(
3427 get_first(video_details, 'lengthSeconds')
3428 or get_first(microformats, 'lengthSeconds')
3429 or parse_duration(search_meta('duration'))) or None
3430
3431 if get_first(video_details, 'isPostLiveDvr'):
3432 self.write_debug('Video is in Post-Live Manifestless mode')
3433 if duration or 0 > 4 * 3600:
3434 self.report_warning(
3435 'The livestream has not finished processing. Only 4 hours of the video can be currently downloaded. '
3436 'This is a known issue and patches are welcome')
3437
3438 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3439 video_id, microformats, video_details, player_responses, player_url, duration)
3440
3441 if not formats:
3442 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3443 self.report_drm(video_id)
3444 pemr = get_first(
3445 playability_statuses,
3446 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3447 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3448 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3449 if subreason:
3450 if subreason == 'The uploader has not made this video available in your country.':
3451 countries = get_first(microformats, 'availableCountries')
3452 if not countries:
3453 regions_allowed = search_meta('regionsAllowed')
3454 countries = regions_allowed.split(',') if regions_allowed else None
3455 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3456 reason += f'. {subreason}'
3457 if reason:
3458 self.raise_no_formats(reason, expected=True)
3459
3460 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3461 if not keywords and webpage:
3462 keywords = [
3463 unescapeHTML(m.group('content'))
3464 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3465 for keyword in keywords:
3466 if keyword.startswith('yt:stretch='):
3467 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3468 if mobj:
3469 # NB: float is intentional for forcing float division
3470 w, h = (float(v) for v in mobj.groups())
3471 if w > 0 and h > 0:
3472 ratio = w / h
3473 for f in formats:
3474 if f.get('vcodec') != 'none':
3475 f['stretched_ratio'] = ratio
3476 break
3477 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3478 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3479 if thumbnail_url:
3480 thumbnails.append({
3481 'url': thumbnail_url,
3482 })
3483 original_thumbnails = thumbnails.copy()
3484
3485 # The best resolution thumbnails sometimes does not appear in the webpage
3486 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3487 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3488 thumbnail_names = [
3489 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3490 # in resolution, these are not the custom thumbnail. So de-prioritize them
3491 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3492 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3493 ]
3494 n_thumbnail_names = len(thumbnail_names)
3495 thumbnails.extend({
3496 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3497 video_id=video_id, name=name, ext=ext,
3498 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3499 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3500 for thumb in thumbnails:
3501 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3502 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3503 self._remove_duplicate_formats(thumbnails)
3504 self._downloader._sort_thumbnails(original_thumbnails)
3505
3506 category = get_first(microformats, 'category') or search_meta('genre')
3507 channel_id = str_or_none(
3508 get_first(video_details, 'channelId')
3509 or get_first(microformats, 'externalChannelId')
3510 or search_meta('channelId'))
3511 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3512
3513 live_content = get_first(video_details, 'isLiveContent')
3514 is_upcoming = get_first(video_details, 'isUpcoming')
3515 if is_live is None:
3516 if is_upcoming or live_content is False:
3517 is_live = False
3518 if is_upcoming is None and (live_content or is_live):
3519 is_upcoming = False
3520 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3521 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3522 if not duration and live_end_time and live_start_time:
3523 duration = live_end_time - live_start_time
3524
3525 if is_live and self.get_param('live_from_start'):
3526 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3527
3528 formats.extend(self._extract_storyboard(player_responses, duration))
3529
3530 # Source is given priority since formats that throttle are given lower source_preference
3531 # When throttling issue is fully fixed, remove this
3532 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3533
3534 info = {
3535 'id': video_id,
3536 'title': video_title,
3537 'formats': formats,
3538 'thumbnails': thumbnails,
3539 # The best thumbnail that we are sure exists. Prevents unnecessary
3540 # URL checking if user don't care about getting the best possible thumbnail
3541 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3542 'description': video_description,
3543 'uploader': get_first(video_details, 'author'),
3544 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3545 'uploader_url': owner_profile_url,
3546 'channel_id': channel_id,
3547 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3548 'duration': duration,
3549 'view_count': int_or_none(
3550 get_first((video_details, microformats), (..., 'viewCount'))
3551 or search_meta('interactionCount')),
3552 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3553 'age_limit': 18 if (
3554 get_first(microformats, 'isFamilySafe') is False
3555 or search_meta('isFamilyFriendly') == 'false'
3556 or search_meta('og:restrictions:age') == '18+') else 0,
3557 'webpage_url': webpage_url,
3558 'categories': [category] if category else None,
3559 'tags': keywords,
3560 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3561 'is_live': is_live,
3562 'was_live': (False if is_live or is_upcoming or live_content is False
3563 else None if is_live is None or is_upcoming is None
3564 else live_content),
3565 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3566 'release_timestamp': live_start_time,
3567 }
3568
3569 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3570 if pctr:
3571 def get_lang_code(track):
3572 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3573 or track.get('languageCode'))
3574
3575 # Converted into dicts to remove duplicates
3576 captions = {
3577 get_lang_code(sub): sub
3578 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3579 translation_languages = {
3580 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3581 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3582
3583 def process_language(container, base_url, lang_code, sub_name, query):
3584 lang_subs = container.setdefault(lang_code, [])
3585 for fmt in self._SUBTITLE_FORMATS:
3586 query.update({
3587 'fmt': fmt,
3588 })
3589 lang_subs.append({
3590 'ext': fmt,
3591 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3592 'name': sub_name,
3593 })
3594
3595 subtitles, automatic_captions = {}, {}
3596 for lang_code, caption_track in captions.items():
3597 base_url = caption_track.get('baseUrl')
3598 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3599 if not base_url:
3600 continue
3601 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3602 if caption_track.get('kind') != 'asr':
3603 if not lang_code:
3604 continue
3605 process_language(
3606 subtitles, base_url, lang_code, lang_name, {})
3607 if not caption_track.get('isTranslatable'):
3608 continue
3609 for trans_code, trans_name in translation_languages.items():
3610 if not trans_code:
3611 continue
3612 orig_trans_code = trans_code
3613 if caption_track.get('kind') != 'asr':
3614 if 'translated_subs' in self._configuration_arg('skip'):
3615 continue
3616 trans_code += f'-{lang_code}'
3617 trans_name += format_field(lang_name, template=' from %s')
3618 # Add an "-orig" label to the original language so that it can be distinguished.
3619 # The subs are returned without "-orig" as well for compatibility
3620 if lang_code == f'a-{orig_trans_code}':
3621 process_language(
3622 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3623 # Setting tlang=lang returns damaged subtitles.
3624 process_language(automatic_captions, base_url, trans_code, trans_name,
3625 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3626 info['automatic_captions'] = automatic_captions
3627 info['subtitles'] = subtitles
3628
3629 parsed_url = compat_urllib_parse_urlparse(url)
3630 for component in [parsed_url.fragment, parsed_url.query]:
3631 query = compat_parse_qs(component)
3632 for k, v in query.items():
3633 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3634 d_k += '_time'
3635 if d_k not in info and k in s_ks:
3636 info[d_k] = parse_duration(query[k][0])
3637
3638 # Youtube Music Auto-generated description
3639 if video_description:
3640 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)
3641 if mobj:
3642 release_year = mobj.group('release_year')
3643 release_date = mobj.group('release_date')
3644 if release_date:
3645 release_date = release_date.replace('-', '')
3646 if not release_year:
3647 release_year = release_date[:4]
3648 info.update({
3649 'album': mobj.group('album'.strip()),
3650 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3651 'track': mobj.group('track').strip(),
3652 'release_date': release_date,
3653 'release_year': int_or_none(release_year),
3654 })
3655
3656 initial_data = None
3657 if webpage:
3658 initial_data = self._extract_yt_initial_variable(
3659 webpage, self._YT_INITIAL_DATA_RE, video_id,
3660 'yt initial data')
3661 if not initial_data:
3662 query = {'videoId': video_id}
3663 query.update(self._get_checkok_params())
3664 initial_data = self._extract_response(
3665 item_id=video_id, ep='next', fatal=False,
3666 ytcfg=master_ytcfg, query=query,
3667 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3668 note='Downloading initial data API JSON')
3669
3670 try: # This will error if there is no livechat
3671 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3672 except (KeyError, IndexError, TypeError):
3673 pass
3674 else:
3675 info.setdefault('subtitles', {})['live_chat'] = [{
3676 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3677 'video_id': video_id,
3678 'ext': 'json',
3679 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3680 }]
3681
3682 if initial_data:
3683 info['chapters'] = (
3684 self._extract_chapters_from_json(initial_data, duration)
3685 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3686 or self._extract_chapters_from_description(video_description, duration)
3687 or None)
3688
3689 contents = traverse_obj(
3690 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3691 expected_type=list, default=[])
3692
3693 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3694 if vpir:
3695 stl = vpir.get('superTitleLink')
3696 if stl:
3697 stl = self._get_text(stl)
3698 if try_get(
3699 vpir,
3700 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3701 info['location'] = stl
3702 else:
3703 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3704 if mobj:
3705 info.update({
3706 'series': mobj.group(1),
3707 'season_number': int(mobj.group(2)),
3708 'episode_number': int(mobj.group(3)),
3709 })
3710 for tlb in (try_get(
3711 vpir,
3712 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3713 list) or []):
3714 tbr = tlb.get('toggleButtonRenderer') or {}
3715 for getter, regex in [(
3716 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3717 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3718 lambda x: x['accessibility'],
3719 lambda x: x['accessibilityData']['accessibilityData'],
3720 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3721 label = (try_get(tbr, getter, dict) or {}).get('label')
3722 if label:
3723 mobj = re.match(regex, label)
3724 if mobj:
3725 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3726 break
3727 sbr_tooltip = try_get(
3728 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3729 if sbr_tooltip:
3730 like_count, dislike_count = sbr_tooltip.split(' / ')
3731 info.update({
3732 'like_count': str_to_int(like_count),
3733 'dislike_count': str_to_int(dislike_count),
3734 })
3735 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3736 if vsir:
3737 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3738 info.update({
3739 'channel': self._get_text(vor, 'title'),
3740 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3741
3742 rows = try_get(
3743 vsir,
3744 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3745 list) or []
3746 multiple_songs = False
3747 for row in rows:
3748 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3749 multiple_songs = True
3750 break
3751 for row in rows:
3752 mrr = row.get('metadataRowRenderer') or {}
3753 mrr_title = mrr.get('title')
3754 if not mrr_title:
3755 continue
3756 mrr_title = self._get_text(mrr, 'title')
3757 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3758 if mrr_title == 'License':
3759 info['license'] = mrr_contents_text
3760 elif not multiple_songs:
3761 if mrr_title == 'Album':
3762 info['album'] = mrr_contents_text
3763 elif mrr_title == 'Artist':
3764 info['artist'] = mrr_contents_text
3765 elif mrr_title == 'Song':
3766 info['track'] = mrr_contents_text
3767
3768 fallbacks = {
3769 'channel': 'uploader',
3770 'channel_id': 'uploader_id',
3771 'channel_url': 'uploader_url',
3772 }
3773
3774 # The upload date for scheduled, live and past live streams / premieres in microformats
3775 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3776 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3777 upload_date = (
3778 unified_strdate(get_first(microformats, 'uploadDate'))
3779 or unified_strdate(search_meta('uploadDate')))
3780 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3781 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3782 info['upload_date'] = upload_date
3783
3784 for to, frm in fallbacks.items():
3785 if not info.get(to):
3786 info[to] = info.get(frm)
3787
3788 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3789 v = info.get(s_k)
3790 if v:
3791 info[d_k] = v
3792
3793 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3794 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3795 is_membersonly = None
3796 is_premium = None
3797 if initial_data and is_private is not None:
3798 is_membersonly = False
3799 is_premium = False
3800 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3801 badge_labels = set()
3802 for content in contents:
3803 if not isinstance(content, dict):
3804 continue
3805 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3806 for badge_label in badge_labels:
3807 if badge_label.lower() == 'members only':
3808 is_membersonly = True
3809 elif badge_label.lower() == 'premium':
3810 is_premium = True
3811 elif badge_label.lower() == 'unlisted':
3812 is_unlisted = True
3813
3814 info['availability'] = self._availability(
3815 is_private=is_private,
3816 needs_premium=is_premium,
3817 needs_subscription=is_membersonly,
3818 needs_auth=info['age_limit'] >= 18,
3819 is_unlisted=None if is_private is None else is_unlisted)
3820
3821 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3822
3823 self.mark_watched(video_id, player_responses)
3824
3825 return info
3826
3827
3828 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3829
3830 @staticmethod
3831 def passthrough_smuggled_data(func):
3832 def _smuggle(entries, smuggled_data):
3833 for entry in entries:
3834 # TODO: Convert URL to music.youtube instead.
3835 # Do we need to passthrough any other smuggled_data?
3836 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3837 yield entry
3838
3839 @functools.wraps(func)
3840 def wrapper(self, url):
3841 url, smuggled_data = unsmuggle_url(url, {})
3842 if self.is_music_url(url):
3843 smuggled_data['is_music_url'] = True
3844 info_dict = func(self, url, smuggled_data)
3845 if smuggled_data and info_dict.get('entries'):
3846 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3847 return info_dict
3848 return wrapper
3849
3850 def _extract_channel_id(self, webpage):
3851 channel_id = self._html_search_meta(
3852 'channelId', webpage, 'channel id', default=None)
3853 if channel_id:
3854 return channel_id
3855 channel_url = self._html_search_meta(
3856 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3857 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3858 'twitter:app:url:googleplay'), webpage, 'channel url')
3859 return self._search_regex(
3860 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3861 channel_url, 'channel id')
3862
3863 @staticmethod
3864 def _extract_basic_item_renderer(item):
3865 # Modified from _extract_grid_item_renderer
3866 known_basic_renderers = (
3867 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3868 )
3869 for key, renderer in item.items():
3870 if not isinstance(renderer, dict):
3871 continue
3872 elif key in known_basic_renderers:
3873 return renderer
3874 elif key.startswith('grid') and key.endswith('Renderer'):
3875 return renderer
3876
3877 def _grid_entries(self, grid_renderer):
3878 for item in grid_renderer['items']:
3879 if not isinstance(item, dict):
3880 continue
3881 renderer = self._extract_basic_item_renderer(item)
3882 if not isinstance(renderer, dict):
3883 continue
3884 title = self._get_text(renderer, 'title')
3885
3886 # playlist
3887 playlist_id = renderer.get('playlistId')
3888 if playlist_id:
3889 yield self.url_result(
3890 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3891 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3892 video_title=title)
3893 continue
3894 # video
3895 video_id = renderer.get('videoId')
3896 if video_id:
3897 yield self._extract_video(renderer)
3898 continue
3899 # channel
3900 channel_id = renderer.get('channelId')
3901 if channel_id:
3902 yield self.url_result(
3903 'https://www.youtube.com/channel/%s' % channel_id,
3904 ie=YoutubeTabIE.ie_key(), video_title=title)
3905 continue
3906 # generic endpoint URL support
3907 ep_url = urljoin('https://www.youtube.com/', try_get(
3908 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3909 compat_str))
3910 if ep_url:
3911 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3912 if ie.suitable(ep_url):
3913 yield self.url_result(
3914 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3915 break
3916
3917 def _music_reponsive_list_entry(self, renderer):
3918 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3919 if video_id:
3920 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3921 ie=YoutubeIE.ie_key(), video_id=video_id)
3922 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3923 if playlist_id:
3924 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3925 if video_id:
3926 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3927 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3928 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3929 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3930 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3931 if browse_id:
3932 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3933 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3934
3935 def _shelf_entries_from_content(self, shelf_renderer):
3936 content = shelf_renderer.get('content')
3937 if not isinstance(content, dict):
3938 return
3939 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3940 if renderer:
3941 # TODO: add support for nested playlists so each shelf is processed
3942 # as separate playlist
3943 # TODO: this includes only first N items
3944 yield from self._grid_entries(renderer)
3945 renderer = content.get('horizontalListRenderer')
3946 if renderer:
3947 # TODO
3948 pass
3949
3950 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3951 ep = try_get(
3952 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3953 compat_str)
3954 shelf_url = urljoin('https://www.youtube.com', ep)
3955 if shelf_url:
3956 # Skipping links to another channels, note that checking for
3957 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3958 # will not work
3959 if skip_channels and '/channels?' in shelf_url:
3960 return
3961 title = self._get_text(shelf_renderer, 'title')
3962 yield self.url_result(shelf_url, video_title=title)
3963 # Shelf may not contain shelf URL, fallback to extraction from content
3964 yield from self._shelf_entries_from_content(shelf_renderer)
3965
3966 def _playlist_entries(self, video_list_renderer):
3967 for content in video_list_renderer['contents']:
3968 if not isinstance(content, dict):
3969 continue
3970 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3971 if not isinstance(renderer, dict):
3972 continue
3973 video_id = renderer.get('videoId')
3974 if not video_id:
3975 continue
3976 yield self._extract_video(renderer)
3977
3978 def _rich_entries(self, rich_grid_renderer):
3979 renderer = try_get(
3980 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3981 video_id = renderer.get('videoId')
3982 if not video_id:
3983 return
3984 yield self._extract_video(renderer)
3985
3986 def _video_entry(self, video_renderer):
3987 video_id = video_renderer.get('videoId')
3988 if video_id:
3989 return self._extract_video(video_renderer)
3990
3991 def _hashtag_tile_entry(self, hashtag_tile_renderer):
3992 url = urljoin('https://youtube.com', traverse_obj(
3993 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
3994 if url:
3995 return self.url_result(
3996 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
3997
3998 def _post_thread_entries(self, post_thread_renderer):
3999 post_renderer = try_get(
4000 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4001 if not post_renderer:
4002 return
4003 # video attachment
4004 video_renderer = try_get(
4005 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4006 video_id = video_renderer.get('videoId')
4007 if video_id:
4008 entry = self._extract_video(video_renderer)
4009 if entry:
4010 yield entry
4011 # playlist attachment
4012 playlist_id = try_get(
4013 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
4014 if playlist_id:
4015 yield self.url_result(
4016 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4017 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4018 # inline video links
4019 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4020 for run in runs:
4021 if not isinstance(run, dict):
4022 continue
4023 ep_url = try_get(
4024 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
4025 if not ep_url:
4026 continue
4027 if not YoutubeIE.suitable(ep_url):
4028 continue
4029 ep_video_id = YoutubeIE._match_id(ep_url)
4030 if video_id == ep_video_id:
4031 continue
4032 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4033
4034 def _post_thread_continuation_entries(self, post_thread_continuation):
4035 contents = post_thread_continuation.get('contents')
4036 if not isinstance(contents, list):
4037 return
4038 for content in contents:
4039 renderer = content.get('backstagePostThreadRenderer')
4040 if not isinstance(renderer, dict):
4041 continue
4042 yield from self._post_thread_entries(renderer)
4043
4044 r''' # unused
4045 def _rich_grid_entries(self, contents):
4046 for content in contents:
4047 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4048 if video_renderer:
4049 entry = self._video_entry(video_renderer)
4050 if entry:
4051 yield entry
4052 '''
4053
4054 def _extract_entries(self, parent_renderer, continuation_list):
4055 # continuation_list is modified in-place with continuation_list = [continuation_token]
4056 continuation_list[:] = [None]
4057 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4058 for content in contents:
4059 if not isinstance(content, dict):
4060 continue
4061 is_renderer = traverse_obj(
4062 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4063 expected_type=dict)
4064 if not is_renderer:
4065 renderer = content.get('richItemRenderer')
4066 if renderer:
4067 for entry in self._rich_entries(renderer):
4068 yield entry
4069 continuation_list[0] = self._extract_continuation(parent_renderer)
4070 continue
4071 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4072 for isr_content in isr_contents:
4073 if not isinstance(isr_content, dict):
4074 continue
4075
4076 known_renderers = {
4077 'playlistVideoListRenderer': self._playlist_entries,
4078 'gridRenderer': self._grid_entries,
4079 'reelShelfRenderer': self._grid_entries,
4080 'shelfRenderer': self._shelf_entries,
4081 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4082 'backstagePostThreadRenderer': self._post_thread_entries,
4083 'videoRenderer': lambda x: [self._video_entry(x)],
4084 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4085 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4086 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4087 }
4088 for key, renderer in isr_content.items():
4089 if key not in known_renderers:
4090 continue
4091 for entry in known_renderers[key](renderer):
4092 if entry:
4093 yield entry
4094 continuation_list[0] = self._extract_continuation(renderer)
4095 break
4096
4097 if not continuation_list[0]:
4098 continuation_list[0] = self._extract_continuation(is_renderer)
4099
4100 if not continuation_list[0]:
4101 continuation_list[0] = self._extract_continuation(parent_renderer)
4102
4103 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4104 continuation_list = [None]
4105 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4106 tab_content = try_get(tab, lambda x: x['content'], dict)
4107 if not tab_content:
4108 return
4109 parent_renderer = (
4110 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4111 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4112 yield from extract_entries(parent_renderer)
4113 continuation = continuation_list[0]
4114
4115 for page_num in itertools.count(1):
4116 if not continuation:
4117 break
4118 headers = self.generate_api_headers(
4119 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4120 response = self._extract_response(
4121 item_id=f'{item_id} page {page_num}',
4122 query=continuation, headers=headers, ytcfg=ytcfg,
4123 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4124
4125 if not response:
4126 break
4127 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4128 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4129 visitor_data = self._extract_visitor_data(response) or visitor_data
4130
4131 known_continuation_renderers = {
4132 'playlistVideoListContinuation': self._playlist_entries,
4133 'gridContinuation': self._grid_entries,
4134 'itemSectionContinuation': self._post_thread_continuation_entries,
4135 'sectionListContinuation': extract_entries, # for feeds
4136 }
4137 continuation_contents = try_get(
4138 response, lambda x: x['continuationContents'], dict) or {}
4139 continuation_renderer = None
4140 for key, value in continuation_contents.items():
4141 if key not in known_continuation_renderers:
4142 continue
4143 continuation_renderer = value
4144 continuation_list = [None]
4145 yield from known_continuation_renderers[key](continuation_renderer)
4146 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4147 break
4148 if continuation_renderer:
4149 continue
4150
4151 known_renderers = {
4152 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4153 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4154 'gridVideoRenderer': (self._grid_entries, 'items'),
4155 'gridChannelRenderer': (self._grid_entries, 'items'),
4156 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4157 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4158 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4159 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4160 }
4161 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4162 continuation_items = try_get(
4163 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4164 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4165 video_items_renderer = None
4166 for key, value in continuation_item.items():
4167 if key not in known_renderers:
4168 continue
4169 video_items_renderer = {known_renderers[key][1]: continuation_items}
4170 continuation_list = [None]
4171 yield from known_renderers[key][0](video_items_renderer)
4172 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4173 break
4174 if video_items_renderer:
4175 continue
4176 break
4177
4178 @staticmethod
4179 def _extract_selected_tab(tabs, fatal=True):
4180 for tab in tabs:
4181 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4182 if renderer.get('selected') is True:
4183 return renderer
4184 else:
4185 if fatal:
4186 raise ExtractorError('Unable to find selected tab')
4187
4188 def _extract_uploader(self, data):
4189 uploader = {}
4190 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4191 owner = try_get(
4192 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4193 if owner:
4194 owner_text = owner.get('text')
4195 uploader['uploader'] = self._search_regex(
4196 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4197 uploader['uploader_id'] = try_get(
4198 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4199 uploader['uploader_url'] = urljoin(
4200 'https://www.youtube.com/',
4201 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4202 return {k: v for k, v in uploader.items() if v is not None}
4203
4204 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4205 playlist_id = title = description = channel_url = channel_name = channel_id = None
4206 tags = []
4207
4208 selected_tab = self._extract_selected_tab(tabs)
4209 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4210 renderer = try_get(
4211 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4212 if renderer:
4213 channel_name = renderer.get('title')
4214 channel_url = renderer.get('channelUrl')
4215 channel_id = renderer.get('externalId')
4216 else:
4217 renderer = try_get(
4218 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4219
4220 if renderer:
4221 title = renderer.get('title')
4222 description = renderer.get('description', '')
4223 playlist_id = channel_id
4224 tags = renderer.get('keywords', '').split()
4225
4226 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4227 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4228 def _get_uncropped(url):
4229 return url_or_none((url or '').split('=')[0] + '=s0')
4230
4231 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4232 if avatar_thumbnails:
4233 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4234 if uncropped_avatar:
4235 avatar_thumbnails.append({
4236 'url': uncropped_avatar,
4237 'id': 'avatar_uncropped',
4238 'preference': 1
4239 })
4240
4241 channel_banners = self._extract_thumbnails(
4242 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4243 for banner in channel_banners:
4244 banner['preference'] = -10
4245
4246 if channel_banners:
4247 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4248 if uncropped_banner:
4249 channel_banners.append({
4250 'url': uncropped_banner,
4251 'id': 'banner_uncropped',
4252 'preference': -5
4253 })
4254
4255 primary_thumbnails = self._extract_thumbnails(
4256 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4257
4258 if playlist_id is None:
4259 playlist_id = item_id
4260
4261 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4262 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4263 if title is None:
4264 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4265 title += format_field(selected_tab, 'title', ' - %s')
4266 title += format_field(selected_tab, 'expandedText', ' - %s')
4267
4268 metadata = {
4269 'playlist_id': playlist_id,
4270 'playlist_title': title,
4271 'playlist_description': description,
4272 'uploader': channel_name,
4273 'uploader_id': channel_id,
4274 'uploader_url': channel_url,
4275 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4276 'tags': tags,
4277 'view_count': self._get_count(playlist_stats, 1),
4278 'availability': self._extract_availability(data),
4279 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4280 'playlist_count': self._get_count(playlist_stats, 0),
4281 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4282 }
4283 if not channel_id:
4284 metadata.update(self._extract_uploader(data))
4285 metadata.update({
4286 'channel': metadata['uploader'],
4287 'channel_id': metadata['uploader_id'],
4288 'channel_url': metadata['uploader_url']})
4289 return self.playlist_result(
4290 self._entries(
4291 selected_tab, playlist_id, ytcfg,
4292 self._extract_account_syncid(ytcfg, data),
4293 self._extract_visitor_data(data, ytcfg)),
4294 **metadata)
4295
4296 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4297 first_id = last_id = response = None
4298 for page_num in itertools.count(1):
4299 videos = list(self._playlist_entries(playlist))
4300 if not videos:
4301 return
4302 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4303 if start >= len(videos):
4304 return
4305 yield from videos[start:]
4306 first_id = first_id or videos[0]['id']
4307 last_id = videos[-1]['id']
4308 watch_endpoint = try_get(
4309 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4310 headers = self.generate_api_headers(
4311 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4312 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4313 query = {
4314 'playlistId': playlist_id,
4315 'videoId': watch_endpoint.get('videoId') or last_id,
4316 'index': watch_endpoint.get('index') or len(videos),
4317 'params': watch_endpoint.get('params') or 'OAE%3D'
4318 }
4319 response = self._extract_response(
4320 item_id='%s page %d' % (playlist_id, page_num),
4321 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4322 check_get_keys='contents'
4323 )
4324 playlist = try_get(
4325 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4326
4327 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4328 title = playlist.get('title') or try_get(
4329 data, lambda x: x['titleText']['simpleText'], compat_str)
4330 playlist_id = playlist.get('playlistId') or item_id
4331
4332 # Delegating everything except mix playlists to regular tab-based playlist URL
4333 playlist_url = urljoin(url, try_get(
4334 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4335 compat_str))
4336
4337 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4338 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4339 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4340
4341 if playlist_url and playlist_url != url and not is_known_unviewable:
4342 return self.url_result(
4343 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4344 video_title=title)
4345
4346 return self.playlist_result(
4347 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4348 playlist_id=playlist_id, playlist_title=title)
4349
4350 def _extract_availability(self, data):
4351 """
4352 Gets the availability of a given playlist/tab.
4353 Note: Unless YouTube tells us explicitly, we do not assume it is public
4354 @param data: response
4355 """
4356 is_private = is_unlisted = None
4357 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4358 badge_labels = self._extract_badges(renderer)
4359
4360 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4361 privacy_dropdown_entries = try_get(
4362 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4363 for renderer_dict in privacy_dropdown_entries:
4364 is_selected = try_get(
4365 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4366 if not is_selected:
4367 continue
4368 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4369 if label:
4370 badge_labels.add(label.lower())
4371 break
4372
4373 for badge_label in badge_labels:
4374 if badge_label == 'unlisted':
4375 is_unlisted = True
4376 elif badge_label == 'private':
4377 is_private = True
4378 elif badge_label == 'public':
4379 is_unlisted = is_private = False
4380 return self._availability(is_private, False, False, False, is_unlisted)
4381
4382 @staticmethod
4383 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4384 sidebar_renderer = try_get(
4385 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4386 for item in sidebar_renderer:
4387 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4388 if renderer:
4389 return renderer
4390
4391 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4392 """
4393 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4394 """
4395 browse_id = params = None
4396 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4397 if not renderer:
4398 return
4399 menu_renderer = try_get(
4400 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4401 for menu_item in menu_renderer:
4402 if not isinstance(menu_item, dict):
4403 continue
4404 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4405 text = try_get(
4406 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4407 if not text or text.lower() != 'show unavailable videos':
4408 continue
4409 browse_endpoint = try_get(
4410 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4411 browse_id = browse_endpoint.get('browseId')
4412 params = browse_endpoint.get('params')
4413 break
4414
4415 headers = self.generate_api_headers(
4416 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4417 visitor_data=self._extract_visitor_data(data, ytcfg))
4418 query = {
4419 'params': params or 'wgYCCAA=',
4420 'browseId': browse_id or 'VL%s' % item_id
4421 }
4422 return self._extract_response(
4423 item_id=item_id, headers=headers, query=query,
4424 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4425 note='Downloading API JSON with unavailable videos')
4426
4427 @functools.cached_property
4428 def skip_webpage(self):
4429 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4430
4431 def _extract_webpage(self, url, item_id, fatal=True):
4432 retries = self.get_param('extractor_retries', 3)
4433 count = -1
4434 webpage = data = last_error = None
4435 while count < retries:
4436 count += 1
4437 # Sometimes youtube returns a webpage with incomplete ytInitialData
4438 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4439 if last_error:
4440 self.report_warning('%s. Retrying ...' % last_error)
4441 try:
4442 webpage = self._download_webpage(
4443 url, item_id,
4444 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4445 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4446 except ExtractorError as e:
4447 if isinstance(e.cause, network_exceptions):
4448 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4449 last_error = error_to_compat_str(e.cause or e.msg)
4450 if count < retries:
4451 continue
4452 if fatal:
4453 raise
4454 self.report_warning(error_to_compat_str(e))
4455 break
4456 else:
4457 try:
4458 self._extract_and_report_alerts(data)
4459 except ExtractorError as e:
4460 if fatal:
4461 raise
4462 self.report_warning(error_to_compat_str(e))
4463 break
4464
4465 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4466 break
4467
4468 last_error = 'Incomplete yt initial data received'
4469 if count >= retries:
4470 if fatal:
4471 raise ExtractorError(last_error)
4472 self.report_warning(last_error)
4473 break
4474
4475 return webpage, data
4476
4477 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4478 """Use if failed to extract ytcfg (and data) from initial webpage"""
4479 if not ytcfg and self.is_authenticated:
4480 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4481 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4482 raise ExtractorError(
4483 f'{msg}. If you are not downloading private content, or '
4484 'your cookies are only for the first account and channel,'
4485 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4486 expected=True)
4487 self.report_warning(msg, only_once=True)
4488
4489 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4490 data = None
4491 if not self.skip_webpage:
4492 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4493 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4494 # Reject webpage data if redirected to home page without explicitly requesting
4495 selected_tab = self._extract_selected_tab(traverse_obj(
4496 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4497 if (url != 'https://www.youtube.com/feed/recommended'
4498 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4499 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4500 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4501 if fatal:
4502 raise ExtractorError(msg, expected=True)
4503 self.report_warning(msg, only_once=True)
4504 if not data:
4505 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4506 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4507 return data, ytcfg
4508
4509 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4510 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4511 resolve_response = self._extract_response(
4512 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4513 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4514 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4515 for ep_key, ep in endpoints.items():
4516 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4517 if params:
4518 return self._extract_response(
4519 item_id=item_id, query=params, ep=ep, headers=headers,
4520 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4521 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4522 err_note = 'Failed to resolve url (does the playlist exist?)'
4523 if fatal:
4524 raise ExtractorError(err_note, expected=True)
4525 self.report_warning(err_note, item_id)
4526
4527 _SEARCH_PARAMS = None
4528
4529 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4530 data = {'query': query}
4531 if params is NO_DEFAULT:
4532 params = self._SEARCH_PARAMS
4533 if params:
4534 data['params'] = params
4535
4536 content_keys = (
4537 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4538 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4539 # ytmusic search
4540 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4541 ('continuationContents', ),
4542 )
4543 display_id = f'query "{query}"'
4544 check_get_keys = tuple({keys[0] for keys in content_keys})
4545 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4546 self._report_playlist_authcheck(ytcfg, fatal=False)
4547
4548 continuation_list = [None]
4549 search = None
4550 for page_num in itertools.count(1):
4551 data.update(continuation_list[0] or {})
4552 headers = self.generate_api_headers(
4553 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4554 search = self._extract_response(
4555 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4556 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4557 slr_contents = traverse_obj(search, *content_keys)
4558 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4559 if not continuation_list[0]:
4560 break
4561
4562
4563 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4564 IE_DESC = 'YouTube Tabs'
4565 _VALID_URL = r'''(?x:
4566 https?://
4567 (?:\w+\.)?
4568 (?:
4569 youtube(?:kids)?\.com|
4570 %(invidious)s
4571 )/
4572 (?:
4573 (?P<channel_type>channel|c|user|browse)/|
4574 (?P<not_channel>
4575 feed/|hashtag/|
4576 (?:playlist|watch)\?.*?\blist=
4577 )|
4578 (?!(?:%(reserved_names)s)\b) # Direct URLs
4579 )
4580 (?P<id>[^/?\#&]+)
4581 )''' % {
4582 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4583 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4584 }
4585 IE_NAME = 'youtube:tab'
4586
4587 _TESTS = [{
4588 'note': 'playlists, multipage',
4589 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4590 'playlist_mincount': 94,
4591 'info_dict': {
4592 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4593 'title': 'Igor Kleiner - Playlists',
4594 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4595 'uploader': 'Igor Kleiner',
4596 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4597 'channel': 'Igor Kleiner',
4598 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4599 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4600 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4601 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4602 'channel_follower_count': int
4603 },
4604 }, {
4605 'note': 'playlists, multipage, different order',
4606 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4607 'playlist_mincount': 94,
4608 'info_dict': {
4609 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4610 'title': 'Igor Kleiner - Playlists',
4611 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4612 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4613 'uploader': 'Igor Kleiner',
4614 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4615 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4616 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4617 'channel': 'Igor Kleiner',
4618 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4619 'channel_follower_count': int
4620 },
4621 }, {
4622 'note': 'playlists, series',
4623 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4624 'playlist_mincount': 5,
4625 'info_dict': {
4626 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4627 'title': '3Blue1Brown - Playlists',
4628 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4629 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4630 'uploader': '3Blue1Brown',
4631 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4632 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4633 'channel': '3Blue1Brown',
4634 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4635 'tags': ['Mathematics'],
4636 'channel_follower_count': int
4637 },
4638 }, {
4639 'note': 'playlists, singlepage',
4640 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4641 'playlist_mincount': 4,
4642 'info_dict': {
4643 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4644 'title': 'ThirstForScience - Playlists',
4645 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4646 'uploader': 'ThirstForScience',
4647 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4648 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4649 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4650 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4651 'tags': 'count:13',
4652 'channel': 'ThirstForScience',
4653 'channel_follower_count': int
4654 }
4655 }, {
4656 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4657 'only_matching': True,
4658 }, {
4659 'note': 'basic, single video playlist',
4660 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4661 'info_dict': {
4662 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4663 'uploader': 'Sergey M.',
4664 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4665 'title': 'youtube-dl public playlist',
4666 'description': '',
4667 'tags': [],
4668 'view_count': int,
4669 'modified_date': '20201130',
4670 'channel': 'Sergey M.',
4671 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4672 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4673 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4674 },
4675 'playlist_count': 1,
4676 }, {
4677 'note': 'empty playlist',
4678 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4679 'info_dict': {
4680 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4681 'uploader': 'Sergey M.',
4682 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4683 'title': 'youtube-dl empty playlist',
4684 'tags': [],
4685 'channel': 'Sergey M.',
4686 'description': '',
4687 'modified_date': '20160902',
4688 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4689 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4690 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4691 },
4692 'playlist_count': 0,
4693 }, {
4694 'note': 'Home tab',
4695 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4696 'info_dict': {
4697 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4698 'title': 'lex will - Home',
4699 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4700 'uploader': 'lex will',
4701 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4702 'channel': 'lex will',
4703 'tags': ['bible', 'history', 'prophesy'],
4704 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4705 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4706 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4707 'channel_follower_count': int
4708 },
4709 'playlist_mincount': 2,
4710 }, {
4711 'note': 'Videos tab',
4712 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4713 'info_dict': {
4714 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4715 'title': 'lex will - Videos',
4716 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4717 'uploader': 'lex will',
4718 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4719 'tags': ['bible', 'history', 'prophesy'],
4720 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4721 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4722 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4723 'channel': 'lex will',
4724 'channel_follower_count': int
4725 },
4726 'playlist_mincount': 975,
4727 }, {
4728 'note': 'Videos tab, sorted by popular',
4729 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4730 'info_dict': {
4731 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4732 'title': 'lex will - Videos',
4733 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4734 'uploader': 'lex will',
4735 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4736 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4737 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4738 'channel': 'lex will',
4739 'tags': ['bible', 'history', 'prophesy'],
4740 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'channel_follower_count': int
4742 },
4743 'playlist_mincount': 199,
4744 }, {
4745 'note': 'Playlists tab',
4746 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4747 'info_dict': {
4748 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4749 'title': 'lex will - Playlists',
4750 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4751 'uploader': 'lex will',
4752 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4753 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4754 'channel': 'lex will',
4755 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4756 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4757 'tags': ['bible', 'history', 'prophesy'],
4758 'channel_follower_count': int
4759 },
4760 'playlist_mincount': 17,
4761 }, {
4762 'note': 'Community tab',
4763 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4764 'info_dict': {
4765 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4766 'title': 'lex will - Community',
4767 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4768 'uploader': 'lex will',
4769 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4770 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4771 'channel': 'lex will',
4772 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4773 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4774 'tags': ['bible', 'history', 'prophesy'],
4775 'channel_follower_count': int
4776 },
4777 'playlist_mincount': 18,
4778 }, {
4779 'note': 'Channels tab',
4780 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4781 'info_dict': {
4782 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4783 'title': 'lex will - Channels',
4784 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4785 'uploader': 'lex will',
4786 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4787 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4788 'channel': 'lex will',
4789 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4790 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4791 'tags': ['bible', 'history', 'prophesy'],
4792 'channel_follower_count': int
4793 },
4794 'playlist_mincount': 12,
4795 }, {
4796 'note': 'Search tab',
4797 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4798 'playlist_mincount': 40,
4799 'info_dict': {
4800 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4801 'title': '3Blue1Brown - Search - linear algebra',
4802 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4803 'uploader': '3Blue1Brown',
4804 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4805 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4806 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4807 'tags': ['Mathematics'],
4808 'channel': '3Blue1Brown',
4809 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4810 'channel_follower_count': int
4811 },
4812 }, {
4813 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4814 'only_matching': True,
4815 }, {
4816 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4817 'only_matching': True,
4818 }, {
4819 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4820 'only_matching': True,
4821 }, {
4822 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4823 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4824 'info_dict': {
4825 'title': '29C3: Not my department',
4826 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4827 'uploader': 'Christiaan008',
4828 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4829 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4830 'tags': [],
4831 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4832 'view_count': int,
4833 'modified_date': '20150605',
4834 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4835 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4836 'channel': 'Christiaan008',
4837 },
4838 'playlist_count': 96,
4839 }, {
4840 'note': 'Large playlist',
4841 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4842 'info_dict': {
4843 'title': 'Uploads from Cauchemar',
4844 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4845 'uploader': 'Cauchemar',
4846 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4847 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4848 'tags': [],
4849 'modified_date': r're:\d{8}',
4850 'channel': 'Cauchemar',
4851 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4852 'view_count': int,
4853 'description': '',
4854 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4855 },
4856 'playlist_mincount': 1123,
4857 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4858 }, {
4859 'note': 'even larger playlist, 8832 videos',
4860 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4861 'only_matching': True,
4862 }, {
4863 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4864 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4865 'info_dict': {
4866 'title': 'Uploads from Interstellar Movie',
4867 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4868 'uploader': 'Interstellar Movie',
4869 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4870 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4871 'tags': [],
4872 'view_count': int,
4873 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4874 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4875 'channel': 'Interstellar Movie',
4876 'description': '',
4877 'modified_date': r're:\d{8}',
4878 },
4879 'playlist_mincount': 21,
4880 }, {
4881 'note': 'Playlist with "show unavailable videos" button',
4882 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4883 'info_dict': {
4884 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4885 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4886 'uploader': 'Phim Siêu Nhân Nhật Bản',
4887 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4888 'view_count': int,
4889 'channel': 'Phim Siêu Nhân Nhật Bản',
4890 'tags': [],
4891 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4892 'description': '',
4893 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4894 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4895 'modified_date': r're:\d{8}',
4896 },
4897 'playlist_mincount': 200,
4898 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4899 }, {
4900 'note': 'Playlist with unavailable videos in page 7',
4901 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4902 'info_dict': {
4903 'title': 'Uploads from BlankTV',
4904 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4905 'uploader': 'BlankTV',
4906 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4907 'channel': 'BlankTV',
4908 'channel_url': 'https://www.youtube.com/c/blanktv',
4909 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4910 'view_count': int,
4911 'tags': [],
4912 'uploader_url': 'https://www.youtube.com/c/blanktv',
4913 'modified_date': r're:\d{8}',
4914 'description': '',
4915 },
4916 'playlist_mincount': 1000,
4917 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4918 }, {
4919 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4920 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4921 'info_dict': {
4922 'title': 'Data Analysis with Dr Mike Pound',
4923 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4924 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4925 'uploader': 'Computerphile',
4926 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4927 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4928 'tags': [],
4929 'view_count': int,
4930 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4931 'channel_url': 'https://www.youtube.com/user/Computerphile',
4932 'channel': 'Computerphile',
4933 },
4934 'playlist_mincount': 11,
4935 }, {
4936 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4937 'only_matching': True,
4938 }, {
4939 'note': 'Playlist URL that does not actually serve a playlist',
4940 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4941 'info_dict': {
4942 'id': 'FqZTN594JQw',
4943 'ext': 'webm',
4944 'title': "Smiley's People 01 detective, Adventure Series, Action",
4945 'uploader': 'STREEM',
4946 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4947 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4948 'upload_date': '20150526',
4949 'license': 'Standard YouTube License',
4950 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4951 'categories': ['People & Blogs'],
4952 'tags': list,
4953 'view_count': int,
4954 'like_count': int,
4955 },
4956 'params': {
4957 'skip_download': True,
4958 },
4959 'skip': 'This video is not available.',
4960 'add_ie': [YoutubeIE.ie_key()],
4961 }, {
4962 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4963 'only_matching': True,
4964 }, {
4965 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4966 'only_matching': True,
4967 }, {
4968 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4969 'info_dict': {
4970 'id': 'GgL890LIznQ', # This will keep changing
4971 'ext': 'mp4',
4972 'title': str,
4973 'uploader': 'Sky News',
4974 'uploader_id': 'skynews',
4975 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
4976 'upload_date': r're:\d{8}',
4977 'description': str,
4978 'categories': ['News & Politics'],
4979 'tags': list,
4980 'like_count': int,
4981 'release_timestamp': 1642502819,
4982 'channel': 'Sky News',
4983 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
4984 'age_limit': 0,
4985 'view_count': int,
4986 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
4987 'playable_in_embed': True,
4988 'release_date': '20220118',
4989 'availability': 'public',
4990 'live_status': 'is_live',
4991 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
4992 'channel_follower_count': int
4993 },
4994 'params': {
4995 'skip_download': True,
4996 },
4997 'expected_warnings': ['Ignoring subtitle tracks found in '],
4998 }, {
4999 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5000 'info_dict': {
5001 'id': 'a48o2S1cPoo',
5002 'ext': 'mp4',
5003 'title': 'The Young Turks - Live Main Show',
5004 'uploader': 'The Young Turks',
5005 'uploader_id': 'TheYoungTurks',
5006 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
5007 'upload_date': '20150715',
5008 'license': 'Standard YouTube License',
5009 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5010 'categories': ['News & Politics'],
5011 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5012 'like_count': int,
5013 },
5014 'params': {
5015 'skip_download': True,
5016 },
5017 'only_matching': True,
5018 }, {
5019 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5020 'only_matching': True,
5021 }, {
5022 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5023 'only_matching': True,
5024 }, {
5025 'note': 'A channel that is not live. Should raise error',
5026 'url': 'https://www.youtube.com/user/numberphile/live',
5027 'only_matching': True,
5028 }, {
5029 'url': 'https://www.youtube.com/feed/trending',
5030 'only_matching': True,
5031 }, {
5032 'url': 'https://www.youtube.com/feed/library',
5033 'only_matching': True,
5034 }, {
5035 'url': 'https://www.youtube.com/feed/history',
5036 'only_matching': True,
5037 }, {
5038 'url': 'https://www.youtube.com/feed/subscriptions',
5039 'only_matching': True,
5040 }, {
5041 'url': 'https://www.youtube.com/feed/watch_later',
5042 'only_matching': True,
5043 }, {
5044 'note': 'Recommended - redirects to home page.',
5045 'url': 'https://www.youtube.com/feed/recommended',
5046 'only_matching': True,
5047 }, {
5048 'note': 'inline playlist with not always working continuations',
5049 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5050 'only_matching': True,
5051 }, {
5052 'url': 'https://www.youtube.com/course',
5053 'only_matching': True,
5054 }, {
5055 'url': 'https://www.youtube.com/zsecurity',
5056 'only_matching': True,
5057 }, {
5058 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5059 'only_matching': True,
5060 }, {
5061 'url': 'https://www.youtube.com/TheYoungTurks/live',
5062 'only_matching': True,
5063 }, {
5064 'url': 'https://www.youtube.com/hashtag/cctv9',
5065 'info_dict': {
5066 'id': 'cctv9',
5067 'title': '#cctv9',
5068 'tags': [],
5069 },
5070 'playlist_mincount': 350,
5071 }, {
5072 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5073 'only_matching': True,
5074 }, {
5075 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5076 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5077 'only_matching': True
5078 }, {
5079 'note': '/browse/ should redirect to /channel/',
5080 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5081 'only_matching': True
5082 }, {
5083 'note': 'VLPL, should redirect to playlist?list=PL...',
5084 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5085 'info_dict': {
5086 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5087 'uploader': 'NoCopyrightSounds',
5088 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5089 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5090 'title': 'NCS Releases',
5091 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5092 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5093 'modified_date': r're:\d{8}',
5094 'view_count': int,
5095 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5096 'tags': [],
5097 'channel': 'NoCopyrightSounds',
5098 },
5099 'playlist_mincount': 166,
5100 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5101 }, {
5102 'note': 'Topic, should redirect to playlist?list=UU...',
5103 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5104 'info_dict': {
5105 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5106 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5107 'title': 'Uploads from Royalty Free Music - Topic',
5108 'uploader': 'Royalty Free Music - Topic',
5109 'tags': [],
5110 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5111 'channel': 'Royalty Free Music - Topic',
5112 'view_count': int,
5113 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5114 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5115 'modified_date': r're:\d{8}',
5116 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5117 'description': '',
5118 },
5119 'expected_warnings': [
5120 'The URL does not have a videos tab',
5121 r'[Uu]navailable videos (are|will be) hidden',
5122 ],
5123 'playlist_mincount': 101,
5124 }, {
5125 'note': 'Topic without a UU playlist',
5126 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5127 'info_dict': {
5128 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5129 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5130 'tags': [],
5131 },
5132 'expected_warnings': [
5133 'the playlist redirect gave error',
5134 ],
5135 'playlist_mincount': 9,
5136 }, {
5137 'note': 'Youtube music Album',
5138 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5139 'info_dict': {
5140 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5141 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5142 'tags': [],
5143 'view_count': int,
5144 'description': '',
5145 'availability': 'unlisted',
5146 'modified_date': r're:\d{8}',
5147 },
5148 'playlist_count': 50,
5149 }, {
5150 'note': 'unlisted single video playlist',
5151 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5152 'info_dict': {
5153 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5154 'uploader': 'colethedj',
5155 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5156 'title': 'yt-dlp unlisted playlist test',
5157 'availability': 'unlisted',
5158 'tags': [],
5159 'modified_date': '20211208',
5160 'channel': 'colethedj',
5161 'view_count': int,
5162 'description': '',
5163 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5164 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5165 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5166 },
5167 'playlist_count': 1,
5168 }, {
5169 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5170 'url': 'https://www.youtube.com/feed/recommended',
5171 'info_dict': {
5172 'id': 'recommended',
5173 'title': 'recommended',
5174 'tags': [],
5175 },
5176 'playlist_mincount': 50,
5177 'params': {
5178 'skip_download': True,
5179 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5180 },
5181 }, {
5182 'note': 'API Fallback: /videos tab, sorted by oldest first',
5183 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5184 'info_dict': {
5185 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5186 'title': 'Cody\'sLab - Videos',
5187 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5188 'uploader': 'Cody\'sLab',
5189 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5190 'channel': 'Cody\'sLab',
5191 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5192 'tags': [],
5193 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5194 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5195 'channel_follower_count': int
5196 },
5197 'playlist_mincount': 650,
5198 'params': {
5199 'skip_download': True,
5200 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5201 },
5202 }, {
5203 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5204 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5205 'info_dict': {
5206 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5207 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5208 'title': 'Uploads from Royalty Free Music - Topic',
5209 'uploader': 'Royalty Free Music - Topic',
5210 'modified_date': r're:\d{8}',
5211 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5212 'description': '',
5213 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5214 'tags': [],
5215 'channel': 'Royalty Free Music - Topic',
5216 'view_count': int,
5217 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5218 },
5219 'expected_warnings': [
5220 'does not have a videos tab',
5221 r'[Uu]navailable videos (are|will be) hidden',
5222 ],
5223 'playlist_mincount': 101,
5224 'params': {
5225 'skip_download': True,
5226 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5227 },
5228 }, {
5229 'note': 'non-standard redirect to regional channel',
5230 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5231 'only_matching': True
5232 }, {
5233 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5234 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5235 'info_dict': {
5236 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5237 'modified_date': '20220407',
5238 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5239 'tags': [],
5240 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5241 'uploader': 'pukkandan',
5242 'availability': 'unlisted',
5243 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5244 'channel': 'pukkandan',
5245 'description': 'Test for collaborative playlist',
5246 'title': 'yt-dlp test - collaborative playlist',
5247 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5248 },
5249 'playlist_mincount': 2
5250 }]
5251
5252 @classmethod
5253 def suitable(cls, url):
5254 return False if YoutubeIE.suitable(url) else super().suitable(url)
5255
5256 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5257
5258 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5259 def _real_extract(self, url, smuggled_data):
5260 item_id = self._match_id(url)
5261 url = compat_urlparse.urlunparse(
5262 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5263 compat_opts = self.get_param('compat_opts', [])
5264
5265 def get_mobj(url):
5266 mobj = self._URL_RE.match(url).groupdict()
5267 mobj.update((k, '') for k, v in mobj.items() if v is None)
5268 return mobj
5269
5270 mobj, redirect_warning = get_mobj(url), None
5271 # Youtube returns incomplete data if tabname is not lower case
5272 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5273 if is_channel:
5274 if smuggled_data.get('is_music_url'):
5275 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5276 item_id = item_id[2:]
5277 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5278 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5279 mdata = self._extract_tab_endpoint(
5280 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5281 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5282 get_all=False, expected_type=compat_str)
5283 if not murl:
5284 raise ExtractorError('Failed to resolve album to playlist')
5285 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5286 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5287 pre = f'https://www.youtube.com/channel/{item_id}'
5288
5289 original_tab_name = tab
5290 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5291 # Home URLs should redirect to /videos/
5292 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5293 'To download only the videos in the home page, add a "/featured" to the URL')
5294 tab = '/videos'
5295
5296 url = ''.join((pre, tab, post))
5297 mobj = get_mobj(url)
5298
5299 # Handle both video/playlist URLs
5300 qs = parse_qs(url)
5301 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5302
5303 if not video_id and mobj['not_channel'].startswith('watch'):
5304 if not playlist_id:
5305 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5306 raise ExtractorError('Unable to recognize tab page')
5307 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5308 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5309 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5310 mobj = get_mobj(url)
5311
5312 if video_id and playlist_id:
5313 if self.get_param('noplaylist'):
5314 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5315 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5316 ie=YoutubeIE.ie_key(), video_id=video_id)
5317 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5318
5319 data, ytcfg = self._extract_data(url, item_id)
5320
5321 # YouTube may provide a non-standard redirect to the regional channel
5322 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5323 redirect_url = traverse_obj(
5324 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5325 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5326 redirect_url = ''.join((
5327 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5328 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5329 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5330
5331 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5332 if tabs:
5333 selected_tab = self._extract_selected_tab(tabs)
5334 selected_tab_name = selected_tab.get('title', '').lower()
5335 if selected_tab_name == 'home':
5336 selected_tab_name = 'featured'
5337 requested_tab_name = mobj['tab'][1:]
5338 if 'no-youtube-channel-redirect' not in compat_opts:
5339 if requested_tab_name == 'live':
5340 # Live tab should have redirected to the video
5341 raise ExtractorError('The channel is not currently live', expected=True)
5342 if requested_tab_name not in ('', selected_tab_name):
5343 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5344 if not original_tab_name:
5345 if item_id[:2] == 'UC':
5346 # Topic channels don't have /videos. Use the equivalent playlist instead
5347 pl_id = f'UU{item_id[2:]}'
5348 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5349 try:
5350 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5351 except ExtractorError:
5352 redirect_warning += ' and the playlist redirect gave error'
5353 else:
5354 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5355 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5356 if selected_tab_name and selected_tab_name != requested_tab_name:
5357 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5358 else:
5359 raise ExtractorError(redirect_warning, expected=True)
5360
5361 if redirect_warning:
5362 self.to_screen(redirect_warning)
5363 self.write_debug(f'Final URL: {url}')
5364
5365 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5366 if 'no-youtube-unavailable-videos' not in compat_opts:
5367 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5368 self._extract_and_report_alerts(data, only_once=True)
5369 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5370 if tabs:
5371 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5372
5373 playlist = traverse_obj(
5374 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5375 if playlist:
5376 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5377
5378 video_id = traverse_obj(
5379 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5380 if video_id:
5381 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5382 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5383 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5384 ie=YoutubeIE.ie_key(), video_id=video_id)
5385
5386 raise ExtractorError('Unable to recognize tab page')
5387
5388
5389 class YoutubePlaylistIE(InfoExtractor):
5390 IE_DESC = 'YouTube playlists'
5391 _VALID_URL = r'''(?x)(?:
5392 (?:https?://)?
5393 (?:\w+\.)?
5394 (?:
5395 (?:
5396 youtube(?:kids)?\.com|
5397 %(invidious)s
5398 )
5399 /.*?\?.*?\blist=
5400 )?
5401 (?P<id>%(playlist_id)s)
5402 )''' % {
5403 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5404 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5405 }
5406 IE_NAME = 'youtube:playlist'
5407 _TESTS = [{
5408 'note': 'issue #673',
5409 'url': 'PLBB231211A4F62143',
5410 'info_dict': {
5411 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5412 'id': 'PLBB231211A4F62143',
5413 'uploader': 'Wickman',
5414 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5415 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5416 'view_count': int,
5417 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5418 'modified_date': r're:\d{8}',
5419 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5420 'channel': 'Wickman',
5421 'tags': [],
5422 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5423 },
5424 'playlist_mincount': 29,
5425 }, {
5426 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5427 'info_dict': {
5428 'title': 'YDL_safe_search',
5429 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5430 },
5431 'playlist_count': 2,
5432 'skip': 'This playlist is private',
5433 }, {
5434 'note': 'embedded',
5435 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5436 'playlist_count': 4,
5437 'info_dict': {
5438 'title': 'JODA15',
5439 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5440 'uploader': 'milan',
5441 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5442 'description': '',
5443 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5444 'tags': [],
5445 'modified_date': '20140919',
5446 'view_count': int,
5447 'channel': 'milan',
5448 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5449 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5450 },
5451 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5452 }, {
5453 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5454 'playlist_mincount': 654,
5455 'info_dict': {
5456 'title': '2018 Chinese New Singles (11/6 updated)',
5457 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5458 'uploader': 'LBK',
5459 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5460 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5461 'channel': 'LBK',
5462 'view_count': int,
5463 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5464 'tags': [],
5465 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5466 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5467 'modified_date': r're:\d{8}',
5468 },
5469 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5470 }, {
5471 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5472 'only_matching': True,
5473 }, {
5474 # music album playlist
5475 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5476 'only_matching': True,
5477 }]
5478
5479 @classmethod
5480 def suitable(cls, url):
5481 if YoutubeTabIE.suitable(url):
5482 return False
5483 from ..utils import parse_qs
5484 qs = parse_qs(url)
5485 if qs.get('v', [None])[0]:
5486 return False
5487 return super().suitable(url)
5488
5489 def _real_extract(self, url):
5490 playlist_id = self._match_id(url)
5491 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5492 url = update_url_query(
5493 'https://www.youtube.com/playlist',
5494 parse_qs(url) or {'list': playlist_id})
5495 if is_music_url:
5496 url = smuggle_url(url, {'is_music_url': True})
5497 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5498
5499
5500 class YoutubeYtBeIE(InfoExtractor):
5501 IE_DESC = 'youtu.be'
5502 _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}
5503 _TESTS = [{
5504 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5505 'info_dict': {
5506 'id': 'yeWKywCrFtk',
5507 'ext': 'mp4',
5508 'title': 'Small Scale Baler and Braiding Rugs',
5509 'uploader': 'Backus-Page House Museum',
5510 'uploader_id': 'backuspagemuseum',
5511 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5512 'upload_date': '20161008',
5513 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5514 'categories': ['Nonprofits & Activism'],
5515 'tags': list,
5516 'like_count': int,
5517 'age_limit': 0,
5518 'playable_in_embed': True,
5519 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5520 'channel': 'Backus-Page House Museum',
5521 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5522 'live_status': 'not_live',
5523 'view_count': int,
5524 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5525 'availability': 'public',
5526 'duration': 59,
5527 },
5528 'params': {
5529 'noplaylist': True,
5530 'skip_download': True,
5531 },
5532 }, {
5533 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5534 'only_matching': True,
5535 }]
5536
5537 def _real_extract(self, url):
5538 mobj = self._match_valid_url(url)
5539 video_id = mobj.group('id')
5540 playlist_id = mobj.group('playlist_id')
5541 return self.url_result(
5542 update_url_query('https://www.youtube.com/watch', {
5543 'v': video_id,
5544 'list': playlist_id,
5545 'feature': 'youtu.be',
5546 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5547
5548
5549 class YoutubeLivestreamEmbedIE(InfoExtractor):
5550 IE_DESC = 'YouTube livestream embeds'
5551 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5552 _TESTS = [{
5553 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5554 'only_matching': True,
5555 }]
5556
5557 def _real_extract(self, url):
5558 channel_id = self._match_id(url)
5559 return self.url_result(
5560 f'https://www.youtube.com/channel/{channel_id}/live',
5561 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5562
5563
5564 class YoutubeYtUserIE(InfoExtractor):
5565 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5566 IE_NAME = 'youtube:user'
5567 _VALID_URL = r'ytuser:(?P<id>.+)'
5568 _TESTS = [{
5569 'url': 'ytuser:phihag',
5570 'only_matching': True,
5571 }]
5572
5573 def _real_extract(self, url):
5574 user_id = self._match_id(url)
5575 return self.url_result(
5576 'https://www.youtube.com/user/%s/videos' % user_id,
5577 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5578
5579
5580 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5581 IE_NAME = 'youtube:favorites'
5582 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5583 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5584 _LOGIN_REQUIRED = True
5585 _TESTS = [{
5586 'url': ':ytfav',
5587 'only_matching': True,
5588 }, {
5589 'url': ':ytfavorites',
5590 'only_matching': True,
5591 }]
5592
5593 def _real_extract(self, url):
5594 return self.url_result(
5595 'https://www.youtube.com/playlist?list=LL',
5596 ie=YoutubeTabIE.ie_key())
5597
5598
5599 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5600 IE_NAME = 'youtube:notif'
5601 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5602 _VALID_URL = r':ytnotif(?:ication)?s?'
5603 _LOGIN_REQUIRED = True
5604 _TESTS = [{
5605 'url': ':ytnotif',
5606 'only_matching': True,
5607 }, {
5608 'url': ':ytnotifications',
5609 'only_matching': True,
5610 }]
5611
5612 def _extract_notification_menu(self, response, continuation_list):
5613 notification_list = traverse_obj(
5614 response,
5615 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5616 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5617 expected_type=list) or []
5618 continuation_list[0] = None
5619 for item in notification_list:
5620 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5621 if entry:
5622 yield entry
5623 continuation = item.get('continuationItemRenderer')
5624 if continuation:
5625 continuation_list[0] = continuation
5626
5627 def _extract_notification_renderer(self, notification):
5628 video_id = traverse_obj(
5629 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5630 url = f'https://www.youtube.com/watch?v={video_id}'
5631 channel_id = None
5632 if not video_id:
5633 browse_ep = traverse_obj(
5634 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5635 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5636 post_id = self._search_regex(
5637 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5638 'post id', default=None)
5639 if not channel_id or not post_id:
5640 return
5641 # The direct /post url redirects to this in the browser
5642 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5643
5644 channel = traverse_obj(
5645 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5646 expected_type=str)
5647 title = self._search_regex(
5648 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5649 'video title', default=None)
5650 if title:
5651 title = title.replace('\xad', '') # remove soft hyphens
5652 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5653 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5654 else None)
5655 return {
5656 '_type': 'url',
5657 'url': url,
5658 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5659 'video_id': video_id,
5660 'title': title,
5661 'channel_id': channel_id,
5662 'channel': channel,
5663 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5664 'upload_date': upload_date,
5665 }
5666
5667 def _notification_menu_entries(self, ytcfg):
5668 continuation_list = [None]
5669 response = None
5670 for page in itertools.count(1):
5671 ctoken = traverse_obj(
5672 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5673 response = self._extract_response(
5674 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5675 ep='notification/get_notification_menu', check_get_keys='actions',
5676 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5677 yield from self._extract_notification_menu(response, continuation_list)
5678 if not continuation_list[0]:
5679 break
5680
5681 def _real_extract(self, url):
5682 display_id = 'notifications'
5683 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5684 self._report_playlist_authcheck(ytcfg)
5685 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5686
5687
5688 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5689 IE_DESC = 'YouTube search'
5690 IE_NAME = 'youtube:search'
5691 _SEARCH_KEY = 'ytsearch'
5692 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5693 _TESTS = [{
5694 'url': 'ytsearch5:youtube-dl test video',
5695 'playlist_count': 5,
5696 'info_dict': {
5697 'id': 'youtube-dl test video',
5698 'title': 'youtube-dl test video',
5699 }
5700 }]
5701
5702
5703 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5704 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5705 _SEARCH_KEY = 'ytsearchdate'
5706 IE_DESC = 'YouTube search, newest videos first'
5707 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5708 _TESTS = [{
5709 'url': 'ytsearchdate5:youtube-dl test video',
5710 'playlist_count': 5,
5711 'info_dict': {
5712 'id': 'youtube-dl test video',
5713 'title': 'youtube-dl test video',
5714 }
5715 }]
5716
5717
5718 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5719 IE_DESC = 'YouTube search URLs with sorting and filter support'
5720 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5721 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5722 _TESTS = [{
5723 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5724 'playlist_mincount': 5,
5725 'info_dict': {
5726 'id': 'youtube-dl test video',
5727 'title': 'youtube-dl test video',
5728 }
5729 }, {
5730 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5731 'playlist_mincount': 5,
5732 'info_dict': {
5733 'id': 'python',
5734 'title': 'python',
5735 }
5736 }, {
5737 'url': 'https://www.youtube.com/results?search_query=%23cats',
5738 'playlist_mincount': 1,
5739 'info_dict': {
5740 'id': '#cats',
5741 'title': '#cats',
5742 'entries': [{
5743 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5744 'title': '#cats',
5745 }],
5746 },
5747 }, {
5748 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5749 'only_matching': True,
5750 }]
5751
5752 def _real_extract(self, url):
5753 qs = parse_qs(url)
5754 query = (qs.get('search_query') or qs.get('q'))[0]
5755 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5756
5757
5758 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5759 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5760 IE_NAME = 'youtube:music:search_url'
5761 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5762 _TESTS = [{
5763 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5764 'playlist_count': 16,
5765 'info_dict': {
5766 'id': 'royalty free music',
5767 'title': 'royalty free music',
5768 }
5769 }, {
5770 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5771 'playlist_mincount': 30,
5772 'info_dict': {
5773 'id': 'royalty free music - songs',
5774 'title': 'royalty free music - songs',
5775 },
5776 'params': {'extract_flat': 'in_playlist'}
5777 }, {
5778 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5779 'playlist_mincount': 30,
5780 'info_dict': {
5781 'id': 'royalty free music - community playlists',
5782 'title': 'royalty free music - community playlists',
5783 },
5784 'params': {'extract_flat': 'in_playlist'}
5785 }]
5786
5787 _SECTIONS = {
5788 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5789 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5790 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5791 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5792 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5793 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5794 }
5795
5796 def _real_extract(self, url):
5797 qs = parse_qs(url)
5798 query = (qs.get('search_query') or qs.get('q'))[0]
5799 params = qs.get('sp', (None,))[0]
5800 if params:
5801 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5802 else:
5803 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5804 params = self._SECTIONS.get(section)
5805 if not params:
5806 section = None
5807 title = join_nonempty(query, section, delim=' - ')
5808 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5809
5810
5811 class YoutubeFeedsInfoExtractor(InfoExtractor):
5812 """
5813 Base class for feed extractors
5814 Subclasses must re-define the _FEED_NAME property.
5815 """
5816 _LOGIN_REQUIRED = True
5817 _FEED_NAME = 'feeds'
5818
5819 def _real_initialize(self):
5820 YoutubeBaseInfoExtractor._check_login_required(self)
5821
5822 @classproperty
5823 def IE_NAME(self):
5824 return f'youtube:{self._FEED_NAME}'
5825
5826 def _real_extract(self, url):
5827 return self.url_result(
5828 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5829
5830
5831 class YoutubeWatchLaterIE(InfoExtractor):
5832 IE_NAME = 'youtube:watchlater'
5833 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5834 _VALID_URL = r':ytwatchlater'
5835 _TESTS = [{
5836 'url': ':ytwatchlater',
5837 'only_matching': True,
5838 }]
5839
5840 def _real_extract(self, url):
5841 return self.url_result(
5842 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5843
5844
5845 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5846 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5847 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5848 _FEED_NAME = 'recommended'
5849 _LOGIN_REQUIRED = False
5850 _TESTS = [{
5851 'url': ':ytrec',
5852 'only_matching': True,
5853 }, {
5854 'url': ':ytrecommended',
5855 'only_matching': True,
5856 }, {
5857 'url': 'https://youtube.com',
5858 'only_matching': True,
5859 }]
5860
5861
5862 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5863 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5864 _VALID_URL = r':ytsub(?:scription)?s?'
5865 _FEED_NAME = 'subscriptions'
5866 _TESTS = [{
5867 'url': ':ytsubs',
5868 'only_matching': True,
5869 }, {
5870 'url': ':ytsubscriptions',
5871 'only_matching': True,
5872 }]
5873
5874
5875 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5876 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5877 _VALID_URL = r':ythis(?:tory)?'
5878 _FEED_NAME = 'history'
5879 _TESTS = [{
5880 'url': ':ythistory',
5881 'only_matching': True,
5882 }]
5883
5884
5885 class YoutubeStoriesIE(InfoExtractor):
5886 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5887 IE_NAME = 'youtube:stories'
5888 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5889 _TESTS = [{
5890 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5891 'only_matching': True,
5892 }]
5893
5894 def _real_extract(self, url):
5895 playlist_id = f'RLTD{self._match_id(url)}'
5896 return self.url_result(
5897 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5898 ie=YoutubeTabIE, video_id=playlist_id)
5899
5900
5901 class YoutubeTruncatedURLIE(InfoExtractor):
5902 IE_NAME = 'youtube:truncated_url'
5903 IE_DESC = False # Do not list
5904 _VALID_URL = r'''(?x)
5905 (?:https?://)?
5906 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5907 (?:watch\?(?:
5908 feature=[a-z_]+|
5909 annotation_id=annotation_[^&]+|
5910 x-yt-cl=[0-9]+|
5911 hl=[^&]*|
5912 t=[0-9]+
5913 )?
5914 |
5915 attribution_link\?a=[^&]+
5916 )
5917 $
5918 '''
5919
5920 _TESTS = [{
5921 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5922 'only_matching': True,
5923 }, {
5924 'url': 'https://www.youtube.com/watch?',
5925 'only_matching': True,
5926 }, {
5927 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5928 'only_matching': True,
5929 }, {
5930 'url': 'https://www.youtube.com/watch?feature=foo',
5931 'only_matching': True,
5932 }, {
5933 'url': 'https://www.youtube.com/watch?hl=en-GB',
5934 'only_matching': True,
5935 }, {
5936 'url': 'https://www.youtube.com/watch?t=2372',
5937 'only_matching': True,
5938 }]
5939
5940 def _real_extract(self, url):
5941 raise ExtractorError(
5942 'Did you forget to quote the URL? Remember that & is a meta '
5943 'character in most shells, so you want to put the URL in quotes, '
5944 'like youtube-dl '
5945 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5946 ' or simply youtube-dl BaW_jenozKc .',
5947 expected=True)
5948
5949
5950 class YoutubeClipIE(InfoExtractor):
5951 IE_NAME = 'youtube:clip'
5952 IE_DESC = False # Do not list
5953 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5954
5955 def _real_extract(self, url):
5956 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5957 return self.url_result(url, 'Generic')
5958
5959
5960 class YoutubeTruncatedIDIE(InfoExtractor):
5961 IE_NAME = 'youtube:truncated_id'
5962 IE_DESC = False # Do not list
5963 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5964
5965 _TESTS = [{
5966 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5967 'only_matching': True,
5968 }]
5969
5970 def _real_extract(self, url):
5971 video_id = self._match_id(url)
5972 raise ExtractorError(
5973 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
5974 expected=True)