]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[extractor] Add dev option `--load-pages`
[yt-dlp.git] / yt_dlp / extractor / youtube.py
... / ...
CommitLineData
1import base64
2import calendar
3import copy
4import datetime
5import hashlib
6import itertools
7import json
8import math
9import os.path
10import random
11import re
12import sys
13import threading
14import time
15import traceback
16
17from .common import InfoExtractor, SearchInfoExtractor
18from ..compat import functools
19from ..compat import (
20 compat_chr,
21 compat_HTTPError,
22 compat_parse_qs,
23 compat_str,
24 compat_urllib_parse_unquote_plus,
25 compat_urllib_parse_urlencode,
26 compat_urllib_parse_urlparse,
27 compat_urlparse,
28)
29from ..jsinterp import JSInterpreter
30from ..utils import (
31 NO_DEFAULT,
32 ExtractorError,
33 bug_reports_message,
34 classproperty,
35 clean_html,
36 datetime_from_str,
37 dict_get,
38 error_to_compat_str,
39 float_or_none,
40 format_field,
41 get_first,
42 int_or_none,
43 is_html,
44 join_nonempty,
45 js_to_json,
46 mimetype2ext,
47 network_exceptions,
48 orderedSet,
49 parse_codecs,
50 parse_count,
51 parse_duration,
52 parse_iso8601,
53 parse_qs,
54 qualities,
55 remove_end,
56 remove_start,
57 smuggle_url,
58 str_or_none,
59 str_to_int,
60 strftime_or_none,
61 traverse_obj,
62 try_get,
63 unescapeHTML,
64 unified_strdate,
65 unified_timestamp,
66 unsmuggle_url,
67 update_url_query,
68 url_or_none,
69 urljoin,
70 variadic,
71)
72
73# any clients starting with _ cannot be explicity requested by the user
74INNERTUBE_CLIENTS = {
75 'web': {
76 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
77 'INNERTUBE_CONTEXT': {
78 'client': {
79 'clientName': 'WEB',
80 'clientVersion': '2.20211221.00.00',
81 }
82 },
83 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
84 },
85 'web_embedded': {
86 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
87 'INNERTUBE_CONTEXT': {
88 'client': {
89 'clientName': 'WEB_EMBEDDED_PLAYER',
90 'clientVersion': '1.20211215.00.01',
91 },
92 },
93 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
94 },
95 'web_music': {
96 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
97 'INNERTUBE_HOST': 'music.youtube.com',
98 'INNERTUBE_CONTEXT': {
99 'client': {
100 'clientName': 'WEB_REMIX',
101 'clientVersion': '1.20211213.00.00',
102 }
103 },
104 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
105 },
106 'web_creator': {
107 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
108 'INNERTUBE_CONTEXT': {
109 'client': {
110 'clientName': 'WEB_CREATOR',
111 'clientVersion': '1.20211220.02.00',
112 }
113 },
114 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
115 },
116 'android': {
117 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
118 'INNERTUBE_CONTEXT': {
119 'client': {
120 'clientName': 'ANDROID',
121 'clientVersion': '16.49',
122 }
123 },
124 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
125 'REQUIRE_JS_PLAYER': False
126 },
127 'android_embedded': {
128 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
129 'INNERTUBE_CONTEXT': {
130 'client': {
131 'clientName': 'ANDROID_EMBEDDED_PLAYER',
132 'clientVersion': '16.49',
133 },
134 },
135 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
136 'REQUIRE_JS_PLAYER': False
137 },
138 'android_music': {
139 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
140 'INNERTUBE_CONTEXT': {
141 'client': {
142 'clientName': 'ANDROID_MUSIC',
143 'clientVersion': '4.57',
144 }
145 },
146 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
147 'REQUIRE_JS_PLAYER': False
148 },
149 'android_creator': {
150 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
151 'INNERTUBE_CONTEXT': {
152 'client': {
153 'clientName': 'ANDROID_CREATOR',
154 'clientVersion': '21.47',
155 },
156 },
157 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
158 'REQUIRE_JS_PLAYER': False
159 },
160 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
161 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
162 'ios': {
163 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
164 'INNERTUBE_CONTEXT': {
165 'client': {
166 'clientName': 'IOS',
167 'clientVersion': '16.46',
168 'deviceModel': 'iPhone14,3',
169 }
170 },
171 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
172 'REQUIRE_JS_PLAYER': False
173 },
174 'ios_embedded': {
175 'INNERTUBE_CONTEXT': {
176 'client': {
177 'clientName': 'IOS_MESSAGES_EXTENSION',
178 'clientVersion': '16.46',
179 'deviceModel': 'iPhone14,3',
180 },
181 },
182 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
183 'REQUIRE_JS_PLAYER': False
184 },
185 'ios_music': {
186 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
187 'INNERTUBE_CONTEXT': {
188 'client': {
189 'clientName': 'IOS_MUSIC',
190 'clientVersion': '4.57',
191 },
192 },
193 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
194 'REQUIRE_JS_PLAYER': False
195 },
196 'ios_creator': {
197 'INNERTUBE_CONTEXT': {
198 'client': {
199 'clientName': 'IOS_CREATOR',
200 'clientVersion': '21.47',
201 },
202 },
203 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
204 'REQUIRE_JS_PLAYER': False
205 },
206 # mweb has 'ultralow' formats
207 # See: https://github.com/yt-dlp/yt-dlp/pull/557
208 'mweb': {
209 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
210 'INNERTUBE_CONTEXT': {
211 'client': {
212 'clientName': 'MWEB',
213 'clientVersion': '2.20211221.01.00',
214 }
215 },
216 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
217 },
218 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
219 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
220 'tv_embedded': {
221 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
222 'INNERTUBE_CONTEXT': {
223 'client': {
224 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
225 'clientVersion': '2.0',
226 },
227 },
228 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
229 },
230}
231
232
233def _split_innertube_client(client_name):
234 variant, *base = client_name.rsplit('.', 1)
235 if base:
236 return variant, base[0], variant
237 base, *variant = client_name.split('_', 1)
238 return client_name, base, variant[0] if variant else None
239
240
241def build_innertube_clients():
242 THIRD_PARTY = {
243 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
244 }
245 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
246 priority = qualities(BASE_CLIENTS[::-1])
247
248 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
249 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
250 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
251 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
252 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
253
254 _, base_client, variant = _split_innertube_client(client)
255 ytcfg['priority'] = 10 * priority(base_client)
256
257 if not variant:
258 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
259 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
260 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
261 embedscreen['priority'] -= 3
262 elif variant == 'embedded':
263 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
264 ytcfg['priority'] -= 2
265 else:
266 ytcfg['priority'] -= 3
267
268
269build_innertube_clients()
270
271
272class YoutubeBaseInfoExtractor(InfoExtractor):
273 """Provide base functions for Youtube extractors"""
274
275 _RESERVED_NAMES = (
276 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
277 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
278 r'browse|oembed|get_video_info|iframe_api|s/player|'
279 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
280
281 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
282
283 # _NETRC_MACHINE = 'youtube'
284
285 # If True it will raise an error if no login info is provided
286 _LOGIN_REQUIRED = False
287
288 _INVIDIOUS_SITES = (
289 # invidious-redirect websites
290 r'(?:www\.)?redirect\.invidious\.io',
291 r'(?:(?:www|dev)\.)?invidio\.us',
292 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
293 r'(?:www\.)?invidious\.pussthecat\.org',
294 r'(?:www\.)?invidious\.zee\.li',
295 r'(?:www\.)?invidious\.ethibox\.fr',
296 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
297 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
298 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
299 # youtube-dl invidious instances list
300 r'(?:(?:www|no)\.)?invidiou\.sh',
301 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
302 r'(?:www\.)?invidious\.kabi\.tk',
303 r'(?:www\.)?invidious\.mastodon\.host',
304 r'(?:www\.)?invidious\.zapashcanon\.fr',
305 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
306 r'(?:www\.)?invidious\.tinfoil-hat\.net',
307 r'(?:www\.)?invidious\.himiko\.cloud',
308 r'(?:www\.)?invidious\.reallyancient\.tech',
309 r'(?:www\.)?invidious\.tube',
310 r'(?:www\.)?invidiou\.site',
311 r'(?:www\.)?invidious\.site',
312 r'(?:www\.)?invidious\.xyz',
313 r'(?:www\.)?invidious\.nixnet\.xyz',
314 r'(?:www\.)?invidious\.048596\.xyz',
315 r'(?:www\.)?invidious\.drycat\.fr',
316 r'(?:www\.)?inv\.skyn3t\.in',
317 r'(?:www\.)?tube\.poal\.co',
318 r'(?:www\.)?tube\.connect\.cafe',
319 r'(?:www\.)?vid\.wxzm\.sx',
320 r'(?:www\.)?vid\.mint\.lgbt',
321 r'(?:www\.)?vid\.puffyan\.us',
322 r'(?:www\.)?yewtu\.be',
323 r'(?:www\.)?yt\.elukerio\.org',
324 r'(?:www\.)?yt\.lelux\.fi',
325 r'(?:www\.)?invidious\.ggc-project\.de',
326 r'(?:www\.)?yt\.maisputain\.ovh',
327 r'(?:www\.)?ytprivate\.com',
328 r'(?:www\.)?invidious\.13ad\.de',
329 r'(?:www\.)?invidious\.toot\.koeln',
330 r'(?:www\.)?invidious\.fdn\.fr',
331 r'(?:www\.)?watch\.nettohikari\.com',
332 r'(?:www\.)?invidious\.namazso\.eu',
333 r'(?:www\.)?invidious\.silkky\.cloud',
334 r'(?:www\.)?invidious\.exonip\.de',
335 r'(?:www\.)?invidious\.riverside\.rocks',
336 r'(?:www\.)?invidious\.blamefran\.net',
337 r'(?:www\.)?invidious\.moomoo\.de',
338 r'(?:www\.)?ytb\.trom\.tf',
339 r'(?:www\.)?yt\.cyberhost\.uk',
340 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
341 r'(?:www\.)?qklhadlycap4cnod\.onion',
342 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
343 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
344 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
345 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
346 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
347 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
348 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
349 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
350 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
351 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
352 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
353 r'(?:www\.)?piped\.kavin\.rocks',
354 r'(?:www\.)?piped\.silkky\.cloud',
355 r'(?:www\.)?piped\.tokhmi\.xyz',
356 r'(?:www\.)?piped\.moomoo\.me',
357 r'(?:www\.)?il\.ax',
358 r'(?:www\.)?piped\.syncpundit\.com',
359 r'(?:www\.)?piped\.mha\.fi',
360 r'(?:www\.)?piped\.mint\.lgbt',
361 r'(?:www\.)?piped\.privacy\.com\.de',
362 )
363
364 def _initialize_consent(self):
365 cookies = self._get_cookies('https://www.youtube.com/')
366 if cookies.get('__Secure-3PSID'):
367 return
368 consent_id = None
369 consent = cookies.get('CONSENT')
370 if consent:
371 if 'YES' in consent.value:
372 return
373 consent_id = self._search_regex(
374 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
375 if not consent_id:
376 consent_id = random.randint(100, 999)
377 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
378
379 def _initialize_pref(self):
380 cookies = self._get_cookies('https://www.youtube.com/')
381 pref_cookie = cookies.get('PREF')
382 pref = {}
383 if pref_cookie:
384 try:
385 pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
386 except ValueError:
387 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
388 pref.update({'hl': 'en', 'tz': 'UTC'})
389 self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
390
391 def _real_initialize(self):
392 self._initialize_pref()
393 self._initialize_consent()
394 self._check_login_required()
395
396 def _check_login_required(self):
397 if self._LOGIN_REQUIRED and not self._cookies_passed:
398 self.raise_login_required('Login details are needed to download this content', method='cookies')
399
400 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+})\s*;'
401 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+})\s*;'
402 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
403
404 def _get_default_ytcfg(self, client='web'):
405 return copy.deepcopy(INNERTUBE_CLIENTS[client])
406
407 def _get_innertube_host(self, client='web'):
408 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
409
410 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
411 # try_get but with fallback to default ytcfg client values when present
412 _func = lambda y: try_get(y, getter, expected_type)
413 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
414
415 def _extract_client_name(self, ytcfg, default_client='web'):
416 return self._ytcfg_get_safe(
417 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
418 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
419
420 def _extract_client_version(self, ytcfg, default_client='web'):
421 return self._ytcfg_get_safe(
422 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
423 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
424
425 def _extract_api_key(self, ytcfg=None, default_client='web'):
426 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
427
428 def _extract_context(self, ytcfg=None, default_client='web'):
429 context = get_first(
430 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
431 # Enforce language and tz for extraction
432 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
433 client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
434 return context
435
436 _SAPISID = None
437
438 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
439 time_now = round(time.time())
440 if self._SAPISID is None:
441 yt_cookies = self._get_cookies('https://www.youtube.com')
442 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
443 # See: https://github.com/yt-dlp/yt-dlp/issues/393
444 sapisid_cookie = dict_get(
445 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
446 if sapisid_cookie and sapisid_cookie.value:
447 self._SAPISID = sapisid_cookie.value
448 self.write_debug('Extracted SAPISID cookie')
449 # SAPISID cookie is required if not already present
450 if not yt_cookies.get('SAPISID'):
451 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
452 self._set_cookie(
453 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
454 else:
455 self._SAPISID = False
456 if not self._SAPISID:
457 return None
458 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
459 sapisidhash = hashlib.sha1(
460 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
461 return f'SAPISIDHASH {time_now}_{sapisidhash}'
462
463 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
464 note='Downloading API JSON', errnote='Unable to download API page',
465 context=None, api_key=None, api_hostname=None, default_client='web'):
466
467 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
468 data.update(query)
469 real_headers = self.generate_api_headers(default_client=default_client)
470 real_headers.update({'content-type': 'application/json'})
471 if headers:
472 real_headers.update(headers)
473 return self._download_json(
474 f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
475 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
476 data=json.dumps(data).encode('utf8'), headers=real_headers,
477 query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
478
479 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
480 data = self._search_regex(
481 (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
482 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
483 if data:
484 return self._parse_json(data, item_id, fatal=fatal)
485
486 @staticmethod
487 def _extract_session_index(*data):
488 """
489 Index of current account in account list.
490 See: https://github.com/yt-dlp/yt-dlp/pull/519
491 """
492 for ytcfg in data:
493 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
494 if session_index is not None:
495 return session_index
496
497 # Deprecated?
498 def _extract_identity_token(self, ytcfg=None, webpage=None):
499 if ytcfg:
500 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
501 if token:
502 return token
503 if webpage:
504 return self._search_regex(
505 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
506 'identity token', default=None, fatal=False)
507
508 @staticmethod
509 def _extract_account_syncid(*args):
510 """
511 Extract syncId required to download private playlists of secondary channels
512 @params response and/or ytcfg
513 """
514 for data in args:
515 # ytcfg includes channel_syncid if on secondary channel
516 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
517 if delegated_sid:
518 return delegated_sid
519 sync_ids = (try_get(
520 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
521 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
522 if len(sync_ids) >= 2 and sync_ids[1]:
523 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
524 # and just "user_syncid||" for primary channel. We only want the channel_syncid
525 return sync_ids[0]
526
527 @staticmethod
528 def _extract_visitor_data(*args):
529 """
530 Extracts visitorData from an API response or ytcfg
531 Appears to be used to track session state
532 """
533 return get_first(
534 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
535 expected_type=str)
536
537 @functools.cached_property
538 def is_authenticated(self):
539 return bool(self._generate_sapisidhash_header())
540
541 def extract_ytcfg(self, video_id, webpage):
542 if not webpage:
543 return {}
544 return self._parse_json(
545 self._search_regex(
546 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
547 default='{}'), video_id, fatal=False) or {}
548
549 def generate_api_headers(
550 self, *, ytcfg=None, account_syncid=None, session_index=None,
551 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
552
553 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
554 headers = {
555 'X-YouTube-Client-Name': compat_str(
556 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
557 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
558 'Origin': origin,
559 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
560 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
561 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
562 }
563 if session_index is None:
564 session_index = self._extract_session_index(ytcfg)
565 if account_syncid or session_index is not None:
566 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
567
568 auth = self._generate_sapisidhash_header(origin)
569 if auth is not None:
570 headers['Authorization'] = auth
571 headers['X-Origin'] = origin
572 return {h: v for h, v in headers.items() if v is not None}
573
574 def _download_ytcfg(self, client, video_id):
575 url = {
576 'web': 'https://www.youtube.com',
577 'web_music': 'https://music.youtube.com',
578 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
579 }.get(client)
580 if not url:
581 return {}
582 webpage = self._download_webpage(
583 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
584 return self.extract_ytcfg(video_id, webpage) or {}
585
586 @staticmethod
587 def _build_api_continuation_query(continuation, ctp=None):
588 query = {
589 'continuation': continuation
590 }
591 # TODO: Inconsistency with clickTrackingParams.
592 # Currently we have a fixed ctp contained within context (from ytcfg)
593 # and a ctp in root query for continuation.
594 if ctp:
595 query['clickTracking'] = {'clickTrackingParams': ctp}
596 return query
597
598 @classmethod
599 def _extract_next_continuation_data(cls, renderer):
600 next_continuation = try_get(
601 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
602 lambda x: x['continuation']['reloadContinuationData']), dict)
603 if not next_continuation:
604 return
605 continuation = next_continuation.get('continuation')
606 if not continuation:
607 return
608 ctp = next_continuation.get('clickTrackingParams')
609 return cls._build_api_continuation_query(continuation, ctp)
610
611 @classmethod
612 def _extract_continuation_ep_data(cls, continuation_ep: dict):
613 if isinstance(continuation_ep, dict):
614 continuation = try_get(
615 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
616 if not continuation:
617 return
618 ctp = continuation_ep.get('clickTrackingParams')
619 return cls._build_api_continuation_query(continuation, ctp)
620
621 @classmethod
622 def _extract_continuation(cls, renderer):
623 next_continuation = cls._extract_next_continuation_data(renderer)
624 if next_continuation:
625 return next_continuation
626
627 contents = []
628 for key in ('contents', 'items'):
629 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
630
631 for content in contents:
632 if not isinstance(content, dict):
633 continue
634 continuation_ep = try_get(
635 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
636 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
637 dict)
638 continuation = cls._extract_continuation_ep_data(continuation_ep)
639 if continuation:
640 return continuation
641
642 @classmethod
643 def _extract_alerts(cls, data):
644 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
645 if not isinstance(alert_dict, dict):
646 continue
647 for alert in alert_dict.values():
648 alert_type = alert.get('type')
649 if not alert_type:
650 continue
651 message = cls._get_text(alert, 'text')
652 if message:
653 yield alert_type, message
654
655 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
656 errors = []
657 warnings = []
658 for alert_type, alert_message in alerts:
659 if alert_type.lower() == 'error' and fatal:
660 errors.append([alert_type, alert_message])
661 else:
662 warnings.append([alert_type, alert_message])
663
664 for alert_type, alert_message in (warnings + errors[:-1]):
665 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
666 if errors:
667 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
668
669 def _extract_and_report_alerts(self, data, *args, **kwargs):
670 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
671
672 def _extract_badges(self, renderer: dict):
673 badges = set()
674 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
675 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
676 if label:
677 badges.add(label.lower())
678 return badges
679
680 @staticmethod
681 def _get_text(data, *path_list, max_runs=None):
682 for path in path_list or [None]:
683 if path is None:
684 obj = [data]
685 else:
686 obj = traverse_obj(data, path, default=[])
687 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
688 obj = [obj]
689 for item in obj:
690 text = try_get(item, lambda x: x['simpleText'], compat_str)
691 if text:
692 return text
693 runs = try_get(item, lambda x: x['runs'], list) or []
694 if not runs and isinstance(item, list):
695 runs = item
696
697 runs = runs[:min(len(runs), max_runs or len(runs))]
698 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
699 if text:
700 return text
701
702 def _get_count(self, data, *path_list):
703 count_text = self._get_text(data, *path_list) or ''
704 count = parse_count(count_text)
705 if count is None:
706 count = str_to_int(
707 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
708 return count
709
710 @staticmethod
711 def _extract_thumbnails(data, *path_list):
712 """
713 Extract thumbnails from thumbnails dict
714 @param path_list: path list to level that contains 'thumbnails' key
715 """
716 thumbnails = []
717 for path in path_list or [()]:
718 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
719 thumbnail_url = url_or_none(thumbnail.get('url'))
720 if not thumbnail_url:
721 continue
722 # Sometimes youtube gives a wrong thumbnail URL. See:
723 # https://github.com/yt-dlp/yt-dlp/issues/233
724 # https://github.com/ytdl-org/youtube-dl/issues/28023
725 if 'maxresdefault' in thumbnail_url:
726 thumbnail_url = thumbnail_url.split('?')[0]
727 thumbnails.append({
728 'url': thumbnail_url,
729 'height': int_or_none(thumbnail.get('height')),
730 'width': int_or_none(thumbnail.get('width')),
731 })
732 return thumbnails
733
734 @staticmethod
735 def extract_relative_time(relative_time_text):
736 """
737 Extracts a relative time from string and converts to dt object
738 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
739 """
740 mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
741 if mobj:
742 start = mobj.group('start')
743 if start:
744 return datetime_from_str(start)
745 try:
746 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
747 except ValueError:
748 return None
749
750 def _extract_time_text(self, renderer, *path_list):
751 """@returns (timestamp, time_text)"""
752 text = self._get_text(renderer, *path_list) or ''
753 dt = self.extract_relative_time(text)
754 timestamp = None
755 if isinstance(dt, datetime.datetime):
756 timestamp = calendar.timegm(dt.timetuple())
757
758 if timestamp is None:
759 timestamp = (
760 unified_timestamp(text) or unified_timestamp(
761 self._search_regex(
762 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
763 text.lower(), 'time text', default=None)))
764
765 if text and timestamp is None:
766 self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
767 return timestamp, text
768
769 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
770 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
771 default_client='web'):
772 response = None
773 last_error = None
774 count = -1
775 retries = self.get_param('extractor_retries', 3)
776 if check_get_keys is None:
777 check_get_keys = []
778 while count < retries:
779 count += 1
780 if last_error:
781 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
782 try:
783 response = self._call_api(
784 ep=ep, fatal=True, headers=headers,
785 video_id=item_id, query=query,
786 context=self._extract_context(ytcfg, default_client),
787 api_key=self._extract_api_key(ytcfg, default_client),
788 api_hostname=api_hostname, default_client=default_client,
789 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
790 except ExtractorError as e:
791 if isinstance(e.cause, network_exceptions):
792 if isinstance(e.cause, compat_HTTPError):
793 first_bytes = e.cause.read(512)
794 if not is_html(first_bytes):
795 yt_error = try_get(
796 self._parse_json(
797 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
798 lambda x: x['error']['message'], compat_str)
799 if yt_error:
800 self._report_alerts([('ERROR', yt_error)], fatal=False)
801 # Downloading page may result in intermittent 5xx HTTP error
802 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
803 # We also want to catch all other network exceptions since errors in later pages can be troublesome
804 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
805 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
806 last_error = error_to_compat_str(e.cause or e.msg)
807 if count < retries:
808 continue
809 if fatal:
810 raise
811 else:
812 self.report_warning(error_to_compat_str(e))
813 return
814
815 else:
816 try:
817 self._extract_and_report_alerts(response, only_once=True)
818 except ExtractorError as e:
819 # YouTube servers may return errors we want to retry on in a 200 OK response
820 # See: https://github.com/yt-dlp/yt-dlp/issues/839
821 if 'unknown error' in e.msg.lower():
822 last_error = e.msg
823 continue
824 if fatal:
825 raise
826 self.report_warning(error_to_compat_str(e))
827 return
828 if not check_get_keys or dict_get(response, check_get_keys):
829 break
830 # Youtube sometimes sends incomplete data
831 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
832 last_error = 'Incomplete data received'
833 if count >= retries:
834 if fatal:
835 raise ExtractorError(last_error)
836 else:
837 self.report_warning(last_error)
838 return
839 return response
840
841 @staticmethod
842 def is_music_url(url):
843 return re.match(r'https?://music\.youtube\.com/', url) is not None
844
845 def _extract_video(self, renderer):
846 video_id = renderer.get('videoId')
847 title = self._get_text(renderer, 'title')
848 description = self._get_text(renderer, 'descriptionSnippet')
849 duration = parse_duration(self._get_text(
850 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
851 if duration is None:
852 duration = parse_duration(self._search_regex(
853 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
854 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
855 video_id, default=None, group='duration'))
856
857 view_count = self._get_count(renderer, 'viewCountText')
858
859 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
860 channel_id = traverse_obj(
861 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
862 expected_type=str, get_all=False)
863 timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
864 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
865 overlay_style = traverse_obj(
866 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
867 get_all=False, expected_type=str)
868 badges = self._extract_badges(renderer)
869 thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
870 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
871 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
872 expected_type=str)) or ''
873 url = f'https://www.youtube.com/watch?v={video_id}'
874 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
875 url = f'https://www.youtube.com/shorts/{video_id}'
876
877 return {
878 '_type': 'url',
879 'ie_key': YoutubeIE.ie_key(),
880 'id': video_id,
881 'url': url,
882 'title': title,
883 'description': description,
884 'duration': duration,
885 'view_count': view_count,
886 'uploader': uploader,
887 'channel_id': channel_id,
888 'thumbnails': thumbnails,
889 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
890 if self._configuration_arg('approximate_date', ie_key='youtubetab')
891 else None),
892 'live_status': ('is_upcoming' if scheduled_timestamp is not None
893 else 'was_live' if 'streamed' in time_text.lower()
894 else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
895 else None),
896 'release_timestamp': scheduled_timestamp,
897 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
898 }
899
900
901class YoutubeIE(YoutubeBaseInfoExtractor):
902 IE_DESC = 'YouTube'
903 _VALID_URL = r"""(?x)^
904 (
905 (?:https?://|//) # http(s):// or protocol-independent URL
906 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
907 (?:www\.)?deturl\.com/www\.youtube\.com|
908 (?:www\.)?pwnyoutube\.com|
909 (?:www\.)?hooktube\.com|
910 (?:www\.)?yourepeat\.com|
911 tube\.majestyc\.net|
912 %(invidious)s|
913 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
914 (?:.*?\#/)? # handle anchor (#/) redirect urls
915 (?: # the various things that can precede the ID:
916 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
917 |(?: # or the v= param in all its forms
918 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
919 (?:\?|\#!?) # the params delimiter ? or # or #!
920 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
921 v=
922 )
923 ))
924 |(?:
925 youtu\.be| # just youtu.be/xxxx
926 vid\.plus| # or vid.plus/xxxx
927 zwearz\.com/watch| # or zwearz.com/watch/xxxx
928 %(invidious)s
929 )/
930 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
931 )
932 )? # all until now is optional -> you can pass the naked ID
933 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
934 (?(1).+)? # if we found the ID, everything can follow
935 (?:\#|$)""" % {
936 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
937 }
938 _PLAYER_INFO_RE = (
939 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
940 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
941 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
942 )
943 _formats = {
944 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
945 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
946 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
947 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
948 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
949 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
950 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
951 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
952 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
953 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
954 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
955 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
956 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
957 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
958 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
959 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
960 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
961 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
962
963
964 # 3D videos
965 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
966 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
967 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
968 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
969 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
970 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
971 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
972
973 # Apple HTTP Live Streaming
974 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
975 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
976 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
977 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
978 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
979 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
980 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
981 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
982
983 # DASH mp4 video
984 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
985 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
986 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
987 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
988 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
989 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
990 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
991 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
992 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
993 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
994 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
995 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
996
997 # Dash mp4 audio
998 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
999 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1000 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1001 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1002 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1003 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1004 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1005
1006 # Dash webm
1007 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1008 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1009 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1010 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1011 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1012 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1013 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1014 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1015 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1016 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1017 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1018 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1019 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1020 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1021 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1022 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1023 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1024 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1025 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1026 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1027 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1028 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1029
1030 # Dash webm audio
1031 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1032 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1033
1034 # Dash webm audio with opus inside
1035 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1036 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1037 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1038
1039 # RTMP (unnamed)
1040 '_rtmp': {'protocol': 'rtmp'},
1041
1042 # av01 video only formats sometimes served with "unknown" codecs
1043 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1044 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1045 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1046 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1047 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1048 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1049 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1050 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1051 }
1052 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1053
1054 _GEO_BYPASS = False
1055
1056 IE_NAME = 'youtube'
1057 _TESTS = [
1058 {
1059 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1060 'info_dict': {
1061 'id': 'BaW_jenozKc',
1062 'ext': 'mp4',
1063 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1064 'uploader': 'Philipp Hagemeister',
1065 'uploader_id': 'phihag',
1066 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1067 'channel': 'Philipp Hagemeister',
1068 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1069 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1070 'upload_date': '20121002',
1071 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1072 'categories': ['Science & Technology'],
1073 'tags': ['youtube-dl'],
1074 'duration': 10,
1075 'view_count': int,
1076 'like_count': int,
1077 'availability': 'public',
1078 'playable_in_embed': True,
1079 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1080 'live_status': 'not_live',
1081 'age_limit': 0,
1082 'start_time': 1,
1083 'end_time': 9,
1084 'channel_follower_count': int
1085 }
1086 },
1087 {
1088 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1089 'note': 'Embed-only video (#1746)',
1090 'info_dict': {
1091 'id': 'yZIXLfi8CZQ',
1092 'ext': 'mp4',
1093 'upload_date': '20120608',
1094 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1095 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1096 'uploader': 'SET India',
1097 'uploader_id': 'setindia',
1098 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1099 'age_limit': 18,
1100 },
1101 'skip': 'Private video',
1102 },
1103 {
1104 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1105 'note': 'Use the first video ID in the URL',
1106 'info_dict': {
1107 'id': 'BaW_jenozKc',
1108 'ext': 'mp4',
1109 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1110 'uploader': 'Philipp Hagemeister',
1111 'uploader_id': 'phihag',
1112 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1113 'channel': 'Philipp Hagemeister',
1114 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1115 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1116 'upload_date': '20121002',
1117 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1118 'categories': ['Science & Technology'],
1119 'tags': ['youtube-dl'],
1120 'duration': 10,
1121 'view_count': int,
1122 'like_count': int,
1123 'availability': 'public',
1124 'playable_in_embed': True,
1125 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1126 'live_status': 'not_live',
1127 'age_limit': 0,
1128 'channel_follower_count': int
1129 },
1130 'params': {
1131 'skip_download': True,
1132 },
1133 },
1134 {
1135 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1136 'note': '256k DASH audio (format 141) via DASH manifest',
1137 'info_dict': {
1138 'id': 'a9LDPn-MO4I',
1139 'ext': 'm4a',
1140 'upload_date': '20121002',
1141 'uploader_id': '8KVIDEO',
1142 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1143 'description': '',
1144 'uploader': '8KVIDEO',
1145 'title': 'UHDTV TEST 8K VIDEO.mp4'
1146 },
1147 'params': {
1148 'youtube_include_dash_manifest': True,
1149 'format': '141',
1150 },
1151 'skip': 'format 141 not served anymore',
1152 },
1153 # DASH manifest with encrypted signature
1154 {
1155 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1156 'info_dict': {
1157 'id': 'IB3lcPjvWLA',
1158 'ext': 'm4a',
1159 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1160 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1161 'duration': 244,
1162 'uploader': 'AfrojackVEVO',
1163 'uploader_id': 'AfrojackVEVO',
1164 'upload_date': '20131011',
1165 'abr': 129.495,
1166 'like_count': int,
1167 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1168 'playable_in_embed': True,
1169 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1170 'view_count': int,
1171 'track': 'The Spark',
1172 'live_status': 'not_live',
1173 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1174 'channel': 'Afrojack',
1175 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1176 'tags': 'count:19',
1177 'availability': 'public',
1178 'categories': ['Music'],
1179 'age_limit': 0,
1180 'alt_title': 'The Spark',
1181 'channel_follower_count': int
1182 },
1183 'params': {
1184 'youtube_include_dash_manifest': True,
1185 'format': '141/bestaudio[ext=m4a]',
1186 },
1187 },
1188 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1189 {
1190 'note': 'Embed allowed age-gate video',
1191 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1192 'info_dict': {
1193 'id': 'HtVdAasjOgU',
1194 'ext': 'mp4',
1195 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1196 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1197 'duration': 142,
1198 'uploader': 'The Witcher',
1199 'uploader_id': 'WitcherGame',
1200 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1201 'upload_date': '20140605',
1202 'age_limit': 18,
1203 'categories': ['Gaming'],
1204 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1205 'availability': 'needs_auth',
1206 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1207 'like_count': int,
1208 'channel': 'The Witcher',
1209 'live_status': 'not_live',
1210 'tags': 'count:17',
1211 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1212 'playable_in_embed': True,
1213 'view_count': int,
1214 'channel_follower_count': int
1215 },
1216 },
1217 {
1218 'note': 'Age-gate video with embed allowed in public site',
1219 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1220 'info_dict': {
1221 'id': 'HsUATh_Nc2U',
1222 'ext': 'mp4',
1223 'title': 'Godzilla 2 (Official Video)',
1224 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1225 'upload_date': '20200408',
1226 'uploader_id': 'FlyingKitty900',
1227 'uploader': 'FlyingKitty',
1228 'age_limit': 18,
1229 'availability': 'needs_auth',
1230 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1231 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1232 'channel': 'FlyingKitty',
1233 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1234 'view_count': int,
1235 'categories': ['Entertainment'],
1236 'live_status': 'not_live',
1237 'tags': ['Flyingkitty', 'godzilla 2'],
1238 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1239 'like_count': int,
1240 'duration': 177,
1241 'playable_in_embed': True,
1242 'channel_follower_count': int
1243 },
1244 },
1245 {
1246 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1247 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1248 'info_dict': {
1249 'id': 'Tq92D6wQ1mg',
1250 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1251 'ext': 'mp4',
1252 'upload_date': '20191228',
1253 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1254 'uploader': 'Projekt Melody',
1255 'description': 'md5:17eccca93a786d51bc67646756894066',
1256 'age_limit': 18,
1257 'like_count': int,
1258 'availability': 'needs_auth',
1259 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1260 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1261 'view_count': int,
1262 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1263 'channel': 'Projekt Melody',
1264 'live_status': 'not_live',
1265 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1266 'playable_in_embed': True,
1267 'categories': ['Entertainment'],
1268 'duration': 106,
1269 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1270 'channel_follower_count': int
1271 },
1272 },
1273 {
1274 'note': 'Non-Agegated non-embeddable video',
1275 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1276 'info_dict': {
1277 'id': 'MeJVWBSsPAY',
1278 'ext': 'mp4',
1279 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1280 'uploader': 'Herr Lurik',
1281 'uploader_id': 'st3in234',
1282 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1283 'upload_date': '20130730',
1284 'track': 'Such mich find mich',
1285 'age_limit': 0,
1286 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1287 'like_count': int,
1288 'playable_in_embed': False,
1289 'creator': 'OOMPH!',
1290 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1291 'view_count': int,
1292 'alt_title': 'Such mich find mich',
1293 'duration': 210,
1294 'channel': 'Herr Lurik',
1295 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1296 'categories': ['Music'],
1297 'availability': 'public',
1298 'uploader_url': 'http://www.youtube.com/user/st3in234',
1299 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1300 'live_status': 'not_live',
1301 'artist': 'OOMPH!',
1302 'channel_follower_count': int
1303 },
1304 },
1305 {
1306 'note': 'Non-bypassable age-gated video',
1307 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1308 'only_matching': True,
1309 },
1310 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1311 # YouTube Red ad is not captured for creator
1312 {
1313 'url': '__2ABJjxzNo',
1314 'info_dict': {
1315 'id': '__2ABJjxzNo',
1316 'ext': 'mp4',
1317 'duration': 266,
1318 'upload_date': '20100430',
1319 'uploader_id': 'deadmau5',
1320 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1321 'creator': 'deadmau5',
1322 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1323 'uploader': 'deadmau5',
1324 'title': 'Deadmau5 - Some Chords (HD)',
1325 'alt_title': 'Some Chords',
1326 'availability': 'public',
1327 'tags': 'count:14',
1328 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1329 'view_count': int,
1330 'live_status': 'not_live',
1331 'channel': 'deadmau5',
1332 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1333 'like_count': int,
1334 'track': 'Some Chords',
1335 'artist': 'deadmau5',
1336 'playable_in_embed': True,
1337 'age_limit': 0,
1338 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1339 'categories': ['Music'],
1340 'album': 'Some Chords',
1341 'channel_follower_count': int
1342 },
1343 'expected_warnings': [
1344 'DASH manifest missing',
1345 ]
1346 },
1347 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1348 {
1349 'url': 'lqQg6PlCWgI',
1350 'info_dict': {
1351 'id': 'lqQg6PlCWgI',
1352 'ext': 'mp4',
1353 'duration': 6085,
1354 'upload_date': '20150827',
1355 'uploader_id': 'olympic',
1356 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1357 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1358 'uploader': 'Olympics',
1359 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1360 'like_count': int,
1361 'release_timestamp': 1343767800,
1362 'playable_in_embed': True,
1363 'categories': ['Sports'],
1364 'release_date': '20120731',
1365 'channel': 'Olympics',
1366 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1367 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1368 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1369 'age_limit': 0,
1370 'availability': 'public',
1371 'live_status': 'was_live',
1372 'view_count': int,
1373 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1374 'channel_follower_count': int
1375 },
1376 'params': {
1377 'skip_download': 'requires avconv',
1378 }
1379 },
1380 # Non-square pixels
1381 {
1382 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1383 'info_dict': {
1384 'id': '_b-2C3KPAM0',
1385 'ext': 'mp4',
1386 'stretched_ratio': 16 / 9.,
1387 'duration': 85,
1388 'upload_date': '20110310',
1389 'uploader_id': 'AllenMeow',
1390 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1391 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
1392 'uploader': '孫ᄋᄅ',
1393 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1394 'playable_in_embed': True,
1395 'channel': '孫ᄋᄅ',
1396 'age_limit': 0,
1397 'tags': 'count:11',
1398 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1399 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1400 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1401 'view_count': int,
1402 'categories': ['People & Blogs'],
1403 'like_count': int,
1404 'live_status': 'not_live',
1405 'availability': 'unlisted',
1406 'channel_follower_count': int
1407 },
1408 },
1409 # url_encoded_fmt_stream_map is empty string
1410 {
1411 'url': 'qEJwOuvDf7I',
1412 'info_dict': {
1413 'id': 'qEJwOuvDf7I',
1414 'ext': 'webm',
1415 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1416 'description': '',
1417 'upload_date': '20150404',
1418 'uploader_id': 'spbelect',
1419 'uploader': 'Наблюдатели Петербурга',
1420 },
1421 'params': {
1422 'skip_download': 'requires avconv',
1423 },
1424 'skip': 'This live event has ended.',
1425 },
1426 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1427 {
1428 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1429 'info_dict': {
1430 'id': 'FIl7x6_3R5Y',
1431 'ext': 'webm',
1432 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1433 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1434 'duration': 220,
1435 'upload_date': '20150625',
1436 'uploader_id': 'dorappi2000',
1437 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1438 'uploader': 'dorappi2000',
1439 'formats': 'mincount:31',
1440 },
1441 'skip': 'not actual anymore',
1442 },
1443 # DASH manifest with segment_list
1444 {
1445 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1446 'md5': '8ce563a1d667b599d21064e982ab9e31',
1447 'info_dict': {
1448 'id': 'CsmdDsKjzN8',
1449 'ext': 'mp4',
1450 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1451 'uploader': 'Airtek',
1452 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1453 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1454 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1455 },
1456 'params': {
1457 'youtube_include_dash_manifest': True,
1458 'format': '135', # bestvideo
1459 },
1460 'skip': 'This live event has ended.',
1461 },
1462 {
1463 # Multifeed videos (multiple cameras), URL is for Main Camera
1464 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1465 'info_dict': {
1466 'id': 'jvGDaLqkpTg',
1467 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1468 'description': 'md5:e03b909557865076822aa169218d6a5d',
1469 },
1470 'playlist': [{
1471 'info_dict': {
1472 'id': 'jvGDaLqkpTg',
1473 'ext': 'mp4',
1474 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1475 'description': 'md5:e03b909557865076822aa169218d6a5d',
1476 'duration': 10643,
1477 'upload_date': '20161111',
1478 'uploader': 'Team PGP',
1479 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1480 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1481 },
1482 }, {
1483 'info_dict': {
1484 'id': '3AKt1R1aDnw',
1485 'ext': 'mp4',
1486 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1487 'description': 'md5:e03b909557865076822aa169218d6a5d',
1488 'duration': 10991,
1489 'upload_date': '20161111',
1490 'uploader': 'Team PGP',
1491 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1492 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1493 },
1494 }, {
1495 'info_dict': {
1496 'id': 'RtAMM00gpVc',
1497 'ext': 'mp4',
1498 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1499 'description': 'md5:e03b909557865076822aa169218d6a5d',
1500 'duration': 10995,
1501 'upload_date': '20161111',
1502 'uploader': 'Team PGP',
1503 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1504 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1505 },
1506 }, {
1507 'info_dict': {
1508 'id': '6N2fdlP3C5U',
1509 'ext': 'mp4',
1510 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1511 'description': 'md5:e03b909557865076822aa169218d6a5d',
1512 'duration': 10990,
1513 'upload_date': '20161111',
1514 'uploader': 'Team PGP',
1515 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1516 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1517 },
1518 }],
1519 'params': {
1520 'skip_download': True,
1521 },
1522 'skip': 'Not multifeed anymore',
1523 },
1524 {
1525 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1526 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1527 'info_dict': {
1528 'id': 'gVfLd0zydlo',
1529 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1530 },
1531 'playlist_count': 2,
1532 'skip': 'Not multifeed anymore',
1533 },
1534 {
1535 'url': 'https://vid.plus/FlRa-iH7PGw',
1536 'only_matching': True,
1537 },
1538 {
1539 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1540 'only_matching': True,
1541 },
1542 {
1543 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1544 # Also tests cut-off URL expansion in video description (see
1545 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1546 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1547 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1548 'info_dict': {
1549 'id': 'lsguqyKfVQg',
1550 'ext': 'mp4',
1551 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1552 'alt_title': 'Dark Walk',
1553 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1554 'duration': 133,
1555 'upload_date': '20151119',
1556 'uploader_id': 'IronSoulElf',
1557 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1558 'uploader': 'IronSoulElf',
1559 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1560 'track': 'Dark Walk',
1561 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1562 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1563 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1564 'categories': ['Film & Animation'],
1565 'view_count': int,
1566 'live_status': 'not_live',
1567 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1568 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1569 'tags': 'count:13',
1570 'availability': 'public',
1571 'channel': 'IronSoulElf',
1572 'playable_in_embed': True,
1573 'like_count': int,
1574 'age_limit': 0,
1575 'channel_follower_count': int
1576 },
1577 'params': {
1578 'skip_download': True,
1579 },
1580 },
1581 {
1582 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1583 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1584 'only_matching': True,
1585 },
1586 {
1587 # Video with yt:stretch=17:0
1588 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1589 'info_dict': {
1590 'id': 'Q39EVAstoRM',
1591 'ext': 'mp4',
1592 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1593 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1594 'upload_date': '20151107',
1595 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1596 'uploader': 'CH GAMER DROID',
1597 },
1598 'params': {
1599 'skip_download': True,
1600 },
1601 'skip': 'This video does not exist.',
1602 },
1603 {
1604 # Video with incomplete 'yt:stretch=16:'
1605 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1606 'only_matching': True,
1607 },
1608 {
1609 # Video licensed under Creative Commons
1610 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1611 'info_dict': {
1612 'id': 'M4gD1WSo5mA',
1613 'ext': 'mp4',
1614 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1615 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1616 'duration': 721,
1617 'upload_date': '20150128',
1618 'uploader_id': 'BerkmanCenter',
1619 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1620 'uploader': 'The Berkman Klein Center for Internet & Society',
1621 'license': 'Creative Commons Attribution license (reuse allowed)',
1622 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1623 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1624 'like_count': int,
1625 'age_limit': 0,
1626 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1627 'channel': 'The Berkman Klein Center for Internet & Society',
1628 'availability': 'public',
1629 'view_count': int,
1630 'categories': ['Education'],
1631 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1632 'live_status': 'not_live',
1633 'playable_in_embed': True,
1634 'channel_follower_count': int
1635 },
1636 'params': {
1637 'skip_download': True,
1638 },
1639 },
1640 {
1641 # Channel-like uploader_url
1642 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1643 'info_dict': {
1644 'id': 'eQcmzGIKrzg',
1645 'ext': 'mp4',
1646 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1647 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1648 'duration': 4060,
1649 'upload_date': '20151120',
1650 'uploader': 'Bernie Sanders',
1651 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1652 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1653 'license': 'Creative Commons Attribution license (reuse allowed)',
1654 'playable_in_embed': True,
1655 'tags': 'count:12',
1656 'like_count': int,
1657 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1658 'age_limit': 0,
1659 'availability': 'public',
1660 'categories': ['News & Politics'],
1661 'channel': 'Bernie Sanders',
1662 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1663 'view_count': int,
1664 'live_status': 'not_live',
1665 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1666 'channel_follower_count': int
1667 },
1668 'params': {
1669 'skip_download': True,
1670 },
1671 },
1672 {
1673 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1674 'only_matching': True,
1675 },
1676 {
1677 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1678 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1679 'only_matching': True,
1680 },
1681 {
1682 # Rental video preview
1683 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1684 'info_dict': {
1685 'id': 'uGpuVWrhIzE',
1686 'ext': 'mp4',
1687 'title': 'Piku - Trailer',
1688 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1689 'upload_date': '20150811',
1690 'uploader': 'FlixMatrix',
1691 'uploader_id': 'FlixMatrixKaravan',
1692 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1693 'license': 'Standard YouTube License',
1694 },
1695 'params': {
1696 'skip_download': True,
1697 },
1698 'skip': 'This video is not available.',
1699 },
1700 {
1701 # YouTube Red video with episode data
1702 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1703 'info_dict': {
1704 'id': 'iqKdEhx-dD4',
1705 'ext': 'mp4',
1706 'title': 'Isolation - Mind Field (Ep 1)',
1707 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1708 'duration': 2085,
1709 'upload_date': '20170118',
1710 'uploader': 'Vsauce',
1711 'uploader_id': 'Vsauce',
1712 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1713 'series': 'Mind Field',
1714 'season_number': 1,
1715 'episode_number': 1,
1716 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1717 'tags': 'count:12',
1718 'view_count': int,
1719 'availability': 'public',
1720 'age_limit': 0,
1721 'channel': 'Vsauce',
1722 'episode': 'Episode 1',
1723 'categories': ['Entertainment'],
1724 'season': 'Season 1',
1725 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1726 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1727 'like_count': int,
1728 'playable_in_embed': True,
1729 'live_status': 'not_live',
1730 'channel_follower_count': int
1731 },
1732 'params': {
1733 'skip_download': True,
1734 },
1735 'expected_warnings': [
1736 'Skipping DASH manifest',
1737 ],
1738 },
1739 {
1740 # The following content has been identified by the YouTube community
1741 # as inappropriate or offensive to some audiences.
1742 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1743 'info_dict': {
1744 'id': '6SJNVb0GnPI',
1745 'ext': 'mp4',
1746 'title': 'Race Differences in Intelligence',
1747 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1748 'duration': 965,
1749 'upload_date': '20140124',
1750 'uploader': 'New Century Foundation',
1751 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1752 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1753 },
1754 'params': {
1755 'skip_download': True,
1756 },
1757 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1758 },
1759 {
1760 # itag 212
1761 'url': '1t24XAntNCY',
1762 'only_matching': True,
1763 },
1764 {
1765 # geo restricted to JP
1766 'url': 'sJL6WA-aGkQ',
1767 'only_matching': True,
1768 },
1769 {
1770 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1771 'only_matching': True,
1772 },
1773 {
1774 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1775 'only_matching': True,
1776 },
1777 {
1778 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1779 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1780 'only_matching': True,
1781 },
1782 {
1783 # DRM protected
1784 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1785 'only_matching': True,
1786 },
1787 {
1788 # Video with unsupported adaptive stream type formats
1789 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1790 'info_dict': {
1791 'id': 'Z4Vy8R84T1U',
1792 'ext': 'mp4',
1793 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1794 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1795 'duration': 433,
1796 'upload_date': '20130923',
1797 'uploader': 'Amelia Putri Harwita',
1798 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1799 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1800 'formats': 'maxcount:10',
1801 },
1802 'params': {
1803 'skip_download': True,
1804 'youtube_include_dash_manifest': False,
1805 },
1806 'skip': 'not actual anymore',
1807 },
1808 {
1809 # Youtube Music Auto-generated description
1810 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1811 'info_dict': {
1812 'id': 'MgNrAu2pzNs',
1813 'ext': 'mp4',
1814 'title': 'Voyeur Girl',
1815 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1816 'upload_date': '20190312',
1817 'uploader': 'Stephen - Topic',
1818 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1819 'artist': 'Stephen',
1820 'track': 'Voyeur Girl',
1821 'album': 'it\'s too much love to know my dear',
1822 'release_date': '20190313',
1823 'release_year': 2019,
1824 'alt_title': 'Voyeur Girl',
1825 'view_count': int,
1826 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1827 'playable_in_embed': True,
1828 'like_count': int,
1829 'categories': ['Music'],
1830 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1831 'channel': 'Stephen',
1832 'availability': 'public',
1833 'creator': 'Stephen',
1834 'duration': 169,
1835 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1836 'age_limit': 0,
1837 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1838 'tags': 'count:11',
1839 'live_status': 'not_live',
1840 'channel_follower_count': int
1841 },
1842 'params': {
1843 'skip_download': True,
1844 },
1845 },
1846 {
1847 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1848 'only_matching': True,
1849 },
1850 {
1851 # invalid -> valid video id redirection
1852 'url': 'DJztXj2GPfl',
1853 'info_dict': {
1854 'id': 'DJztXj2GPfk',
1855 'ext': 'mp4',
1856 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1857 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1858 'upload_date': '20090125',
1859 'uploader': 'Prochorowka',
1860 'uploader_id': 'Prochorowka',
1861 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1862 'artist': 'Panjabi MC',
1863 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1864 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1865 },
1866 'params': {
1867 'skip_download': True,
1868 },
1869 'skip': 'Video unavailable',
1870 },
1871 {
1872 # empty description results in an empty string
1873 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1874 'info_dict': {
1875 'id': 'x41yOUIvK2k',
1876 'ext': 'mp4',
1877 'title': 'IMG 3456',
1878 'description': '',
1879 'upload_date': '20170613',
1880 'uploader_id': 'ElevageOrVert',
1881 'uploader': 'ElevageOrVert',
1882 'view_count': int,
1883 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
1884 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
1885 'like_count': int,
1886 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
1887 'tags': [],
1888 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
1889 'availability': 'public',
1890 'age_limit': 0,
1891 'categories': ['Pets & Animals'],
1892 'duration': 7,
1893 'playable_in_embed': True,
1894 'live_status': 'not_live',
1895 'channel': 'ElevageOrVert',
1896 'channel_follower_count': int
1897 },
1898 'params': {
1899 'skip_download': True,
1900 },
1901 },
1902 {
1903 # with '};' inside yt initial data (see [1])
1904 # see [2] for an example with '};' inside ytInitialPlayerResponse
1905 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1906 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1907 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1908 'info_dict': {
1909 'id': 'CHqg6qOn4no',
1910 'ext': 'mp4',
1911 'title': 'Part 77 Sort a list of simple types in c#',
1912 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1913 'upload_date': '20130831',
1914 'uploader_id': 'kudvenkat',
1915 'uploader': 'kudvenkat',
1916 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
1917 'like_count': int,
1918 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
1919 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
1920 'live_status': 'not_live',
1921 'categories': ['Education'],
1922 'availability': 'public',
1923 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
1924 'tags': 'count:12',
1925 'playable_in_embed': True,
1926 'age_limit': 0,
1927 'view_count': int,
1928 'duration': 522,
1929 'channel': 'kudvenkat',
1930 'channel_follower_count': int
1931 },
1932 'params': {
1933 'skip_download': True,
1934 },
1935 },
1936 {
1937 # another example of '};' in ytInitialData
1938 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1939 'only_matching': True,
1940 },
1941 {
1942 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1943 'only_matching': True,
1944 },
1945 {
1946 # https://github.com/ytdl-org/youtube-dl/pull/28094
1947 'url': 'OtqTfy26tG0',
1948 'info_dict': {
1949 'id': 'OtqTfy26tG0',
1950 'ext': 'mp4',
1951 'title': 'Burn Out',
1952 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1953 'upload_date': '20141120',
1954 'uploader': 'The Cinematic Orchestra - Topic',
1955 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1956 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1957 'artist': 'The Cinematic Orchestra',
1958 'track': 'Burn Out',
1959 'album': 'Every Day',
1960 'like_count': int,
1961 'live_status': 'not_live',
1962 'alt_title': 'Burn Out',
1963 'duration': 614,
1964 'age_limit': 0,
1965 'view_count': int,
1966 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1967 'creator': 'The Cinematic Orchestra',
1968 'channel': 'The Cinematic Orchestra',
1969 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
1970 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1971 'availability': 'public',
1972 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
1973 'categories': ['Music'],
1974 'playable_in_embed': True,
1975 'channel_follower_count': int
1976 },
1977 'params': {
1978 'skip_download': True,
1979 },
1980 },
1981 {
1982 # controversial video, only works with bpctr when authenticated with cookies
1983 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1984 'only_matching': True,
1985 },
1986 {
1987 # controversial video, requires bpctr/contentCheckOk
1988 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1989 'info_dict': {
1990 'id': 'SZJvDhaSDnc',
1991 'ext': 'mp4',
1992 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1993 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1994 'uploader': 'CBS Mornings',
1995 'uploader_id': 'CBSThisMorning',
1996 'upload_date': '20140716',
1997 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
1998 'duration': 170,
1999 'categories': ['News & Politics'],
2000 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2001 'view_count': int,
2002 'channel': 'CBS Mornings',
2003 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2004 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2005 'age_limit': 18,
2006 'availability': 'needs_auth',
2007 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2008 'like_count': int,
2009 'live_status': 'not_live',
2010 'playable_in_embed': True,
2011 'channel_follower_count': int
2012 }
2013 },
2014 {
2015 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2016 'url': 'cBvYw8_A0vQ',
2017 'info_dict': {
2018 'id': 'cBvYw8_A0vQ',
2019 'ext': 'mp4',
2020 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2021 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2022 'upload_date': '20201120',
2023 'uploader': 'Walk around Japan',
2024 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2026 'duration': 1456,
2027 'categories': ['Travel & Events'],
2028 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2029 'view_count': int,
2030 'channel': 'Walk around Japan',
2031 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2032 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2033 'age_limit': 0,
2034 'availability': 'public',
2035 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2036 'live_status': 'not_live',
2037 'playable_in_embed': True,
2038 'channel_follower_count': int
2039 },
2040 'params': {
2041 'skip_download': True,
2042 },
2043 }, {
2044 # Has multiple audio streams
2045 'url': 'WaOKSUlf4TM',
2046 'only_matching': True
2047 }, {
2048 # Requires Premium: has format 141 when requested using YTM url
2049 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2050 'only_matching': True
2051 }, {
2052 # multiple subtitles with same lang_code
2053 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2054 'only_matching': True,
2055 }, {
2056 # Force use android client fallback
2057 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2058 'info_dict': {
2059 'id': 'YOelRv7fMxY',
2060 'title': 'DIGGING A SECRET TUNNEL Part 1',
2061 'ext': '3gp',
2062 'upload_date': '20210624',
2063 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2064 'uploader': 'colinfurze',
2065 'uploader_id': 'colinfurze',
2066 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2067 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2068 'duration': 596,
2069 'categories': ['Entertainment'],
2070 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2071 'view_count': int,
2072 'channel': 'colinfurze',
2073 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2074 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2075 'age_limit': 0,
2076 'availability': 'public',
2077 'like_count': int,
2078 'live_status': 'not_live',
2079 'playable_in_embed': True,
2080 'channel_follower_count': int
2081 },
2082 'params': {
2083 'format': '17', # 3gp format available on android
2084 'extractor_args': {'youtube': {'player_client': ['android']}},
2085 },
2086 },
2087 {
2088 # Skip download of additional client configs (remix client config in this case)
2089 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2090 'only_matching': True,
2091 'params': {
2092 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2093 },
2094 }, {
2095 # shorts
2096 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2097 'only_matching': True,
2098 }, {
2099 'note': 'Storyboards',
2100 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2101 'info_dict': {
2102 'id': '5KLPxDtMqe8',
2103 'ext': 'mhtml',
2104 'format_id': 'sb0',
2105 'title': 'Your Brain is Plastic',
2106 'uploader_id': 'scishow',
2107 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2108 'upload_date': '20140324',
2109 'uploader': 'SciShow',
2110 'like_count': int,
2111 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2112 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2113 'view_count': int,
2114 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2115 'playable_in_embed': True,
2116 'tags': 'count:12',
2117 'uploader_url': 'http://www.youtube.com/user/scishow',
2118 'availability': 'public',
2119 'channel': 'SciShow',
2120 'live_status': 'not_live',
2121 'duration': 248,
2122 'categories': ['Education'],
2123 'age_limit': 0,
2124 'channel_follower_count': int
2125 }, 'params': {'format': 'mhtml', 'skip_download': True}
2126 }, {
2127 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2128 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2129 'info_dict': {
2130 'id': '2NUZ8W2llS4',
2131 'ext': 'mp4',
2132 'title': 'The NP that test your phone performance 🙂',
2133 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2134 'uploader': 'Leon Nguyen',
2135 'uploader_id': 'VNSXIII',
2136 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2137 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2138 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2139 'duration': 21,
2140 'view_count': int,
2141 'age_limit': 0,
2142 'categories': ['Gaming'],
2143 'tags': 'count:23',
2144 'playable_in_embed': True,
2145 'live_status': 'not_live',
2146 'upload_date': '20220103',
2147 'like_count': int,
2148 'availability': 'public',
2149 'channel': 'Leon Nguyen',
2150 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2151 'channel_follower_count': int
2152 }
2153 }, {
2154 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2155 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2156 'info_dict': {
2157 'id': 'mzZzzBU6lrM',
2158 'ext': 'mp4',
2159 'title': 'I Met GeorgeNotFound In Real Life...',
2160 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2161 'uploader': 'Quackity',
2162 'uploader_id': 'QuackityHQ',
2163 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2164 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2165 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2166 'duration': 955,
2167 'view_count': int,
2168 'age_limit': 0,
2169 'categories': ['Entertainment'],
2170 'tags': 'count:26',
2171 'playable_in_embed': True,
2172 'live_status': 'not_live',
2173 'release_timestamp': 1641172509,
2174 'release_date': '20220103',
2175 'upload_date': '20220103',
2176 'like_count': int,
2177 'availability': 'public',
2178 'channel': 'Quackity',
2179 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2180 'channel_follower_count': int
2181 }
2182 },
2183 { # continuous livestream. Microformat upload date should be preferred.
2184 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2185 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2186 'info_dict': {
2187 'id': 'kgx4WGK0oNU',
2188 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2189 'ext': 'mp4',
2190 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2191 'availability': 'public',
2192 'age_limit': 0,
2193 'release_timestamp': 1637975704,
2194 'upload_date': '20210619',
2195 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2196 'live_status': 'is_live',
2197 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2198 'uploader': '阿鲍Abao',
2199 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2200 'channel': 'Abao in Tokyo',
2201 'channel_follower_count': int,
2202 'release_date': '20211127',
2203 'tags': 'count:39',
2204 'categories': ['People & Blogs'],
2205 'like_count': int,
2206 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2207 'view_count': int,
2208 'playable_in_embed': True,
2209 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2210 },
2211 'params': {'skip_download': True}
2212 }, {
2213 # Story. Requires specific player params to work.
2214 # Note: stories get removed after some period of time
2215 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2216 'info_dict': {
2217 'id': 'vv8qTUWmulI',
2218 'ext': 'mp4',
2219 'availability': 'unlisted',
2220 'view_count': int,
2221 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2222 'upload_date': '20220526',
2223 'categories': ['Education'],
2224 'title': 'Story',
2225 'channel': 'IT\'S HISTORY',
2226 'description': '',
2227 'uploader_id': 'BlastfromthePast',
2228 'duration': 12,
2229 'uploader': 'IT\'S HISTORY',
2230 'playable_in_embed': True,
2231 'age_limit': 0,
2232 'live_status': 'not_live',
2233 'tags': [],
2234 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2235 'uploader_url': 'http://www.youtube.com/user/BlastfromthePast',
2236 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2237 }
2238 }, {
2239 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2240 'info_dict': {
2241 'id': 'tjjjtzRLHvA',
2242 'ext': 'mp4',
2243 'title': 'ハッシュタグ無し };if window.ytcsi',
2244 'upload_date': '20220323',
2245 'like_count': int,
2246 'availability': 'unlisted',
2247 'channel': 'nao20010128nao',
2248 'thumbnail': 'https://i.ytimg.com/vi_webp/tjjjtzRLHvA/maxresdefault.webp',
2249 'age_limit': 0,
2250 'uploader': 'nao20010128nao',
2251 'uploader_id': 'nao20010128nao',
2252 'categories': ['Music'],
2253 'view_count': int,
2254 'description': '',
2255 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2256 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2257 'live_status': 'not_live',
2258 'playable_in_embed': True,
2259 'channel_follower_count': int,
2260 'duration': 6,
2261 'tags': [],
2262 'uploader_url': 'http://www.youtube.com/user/nao20010128nao',
2263 }
2264 }
2265 ]
2266
2267 @classmethod
2268 def suitable(cls, url):
2269 from ..utils import parse_qs
2270
2271 qs = parse_qs(url)
2272 if qs.get('list', [None])[0]:
2273 return False
2274 return super().suitable(url)
2275
2276 def __init__(self, *args, **kwargs):
2277 super().__init__(*args, **kwargs)
2278 self._code_cache = {}
2279 self._player_cache = {}
2280
2281 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
2282 lock = threading.Lock()
2283
2284 is_live = True
2285 start_time = time.time()
2286 formats = [f for f in formats if f.get('is_from_start')]
2287
2288 def refetch_manifest(format_id, delay):
2289 nonlocal formats, start_time, is_live
2290 if time.time() <= start_time + delay:
2291 return
2292
2293 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2294 video_details = traverse_obj(
2295 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2296 microformats = traverse_obj(
2297 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2298 expected_type=dict, default=[])
2299 _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
2300 start_time = time.time()
2301
2302 def mpd_feed(format_id, delay):
2303 """
2304 @returns (manifest_url, manifest_stream_number, is_live) or None
2305 """
2306 with lock:
2307 refetch_manifest(format_id, delay)
2308
2309 f = next((f for f in formats if f['format_id'] == format_id), None)
2310 if not f:
2311 if not is_live:
2312 self.to_screen(f'{video_id}: Video is no longer live')
2313 else:
2314 self.report_warning(
2315 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2316 return None
2317 return f['manifest_url'], f['manifest_stream_number'], is_live
2318
2319 for f in formats:
2320 f['is_live'] = True
2321 f['protocol'] = 'http_dash_segments_generator'
2322 f['fragments'] = functools.partial(
2323 self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
2324
2325 def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
2326 FETCH_SPAN, MAX_DURATION = 5, 432000
2327
2328 mpd_url, stream_number, is_live = None, None, True
2329
2330 begin_index = 0
2331 download_start_time = ctx.get('start') or time.time()
2332
2333 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2334 if lack_early_segments:
2335 self.report_warning(bug_reports_message(
2336 'Starting download from the last 120 hours of the live stream since '
2337 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2338 lack_early_segments = True
2339
2340 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2341 fragments, fragment_base_url = None, None
2342
2343 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2344 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2345 # Obtain from MPD's maximum seq value
2346 old_mpd_url = mpd_url
2347 last_error = ctx.pop('last_error', None)
2348 expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
2349 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2350 or (mpd_url, stream_number, False))
2351 if not refresh_sequence:
2352 if expire_fast and not is_live:
2353 return False, last_seq
2354 elif old_mpd_url == mpd_url:
2355 return True, last_seq
2356 try:
2357 fmts, _ = self._extract_mpd_formats_and_subtitles(
2358 mpd_url, None, note=False, errnote=False, fatal=False)
2359 except ExtractorError:
2360 fmts = None
2361 if not fmts:
2362 no_fragment_score += 2
2363 return False, last_seq
2364 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2365 fragments = fmt_info['fragments']
2366 fragment_base_url = fmt_info['fragment_base_url']
2367 assert fragment_base_url
2368
2369 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2370 return True, _last_seq
2371
2372 while is_live:
2373 fetch_time = time.time()
2374 if no_fragment_score > 30:
2375 return
2376 if last_segment_url:
2377 # Obtain from "X-Head-Seqnum" header value from each segment
2378 try:
2379 urlh = self._request_webpage(
2380 last_segment_url, None, note=False, errnote=False, fatal=False)
2381 except ExtractorError:
2382 urlh = None
2383 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2384 if last_seq is None:
2385 no_fragment_score += 2
2386 last_segment_url = None
2387 continue
2388 else:
2389 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2390 no_fragment_score += 2
2391 if not should_continue:
2392 continue
2393
2394 if known_idx > last_seq:
2395 last_segment_url = None
2396 continue
2397
2398 last_seq += 1
2399
2400 if begin_index < 0 and known_idx < 0:
2401 # skip from the start when it's negative value
2402 known_idx = last_seq + begin_index
2403 if lack_early_segments:
2404 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2405 try:
2406 for idx in range(known_idx, last_seq):
2407 # do not update sequence here or you'll get skipped some part of it
2408 should_continue, _ = _extract_sequence_from_mpd(False, False)
2409 if not should_continue:
2410 known_idx = idx - 1
2411 raise ExtractorError('breaking out of outer loop')
2412 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2413 yield {
2414 'url': last_segment_url,
2415 }
2416 if known_idx == last_seq:
2417 no_fragment_score += 5
2418 else:
2419 no_fragment_score = 0
2420 known_idx = last_seq
2421 except ExtractorError:
2422 continue
2423
2424 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2425
2426 def _extract_player_url(self, *ytcfgs, webpage=None):
2427 player_url = traverse_obj(
2428 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2429 get_all=False, expected_type=compat_str)
2430 if not player_url:
2431 return
2432 return urljoin('https://www.youtube.com', player_url)
2433
2434 def _download_player_url(self, video_id, fatal=False):
2435 res = self._download_webpage(
2436 'https://www.youtube.com/iframe_api',
2437 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2438 if res:
2439 player_version = self._search_regex(
2440 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2441 if player_version:
2442 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2443
2444 def _signature_cache_id(self, example_sig):
2445 """ Return a string representation of a signature """
2446 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
2447
2448 @classmethod
2449 def _extract_player_info(cls, player_url):
2450 for player_re in cls._PLAYER_INFO_RE:
2451 id_m = re.search(player_re, player_url)
2452 if id_m:
2453 break
2454 else:
2455 raise ExtractorError('Cannot identify player %r' % player_url)
2456 return id_m.group('id')
2457
2458 def _load_player(self, video_id, player_url, fatal=True):
2459 player_id = self._extract_player_info(player_url)
2460 if player_id not in self._code_cache:
2461 code = self._download_webpage(
2462 player_url, video_id, fatal=fatal,
2463 note='Downloading player ' + player_id,
2464 errnote='Download of %s failed' % player_url)
2465 if code:
2466 self._code_cache[player_id] = code
2467 return self._code_cache.get(player_id)
2468
2469 def _extract_signature_function(self, video_id, player_url, example_sig):
2470 player_id = self._extract_player_info(player_url)
2471
2472 # Read from filesystem cache
2473 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2474 assert os.path.basename(func_id) == func_id
2475
2476 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
2477 if cache_spec is not None:
2478 return lambda s: ''.join(s[i] for i in cache_spec)
2479
2480 code = self._load_player(video_id, player_url)
2481 if code:
2482 res = self._parse_sig_js(code)
2483
2484 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2485 cache_res = res(test_string)
2486 cache_spec = [ord(c) for c in cache_res]
2487
2488 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
2489 return res
2490
2491 def _print_sig_code(self, func, example_sig):
2492 if not self.get_param('youtube_print_sig_code'):
2493 return
2494
2495 def gen_sig_code(idxs):
2496 def _genslice(start, end, step):
2497 starts = '' if start == 0 else str(start)
2498 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2499 steps = '' if step == 1 else (':%d' % step)
2500 return f's[{starts}{ends}{steps}]'
2501
2502 step = None
2503 # Quelch pyflakes warnings - start will be set when step is set
2504 start = '(Never used)'
2505 for i, prev in zip(idxs[1:], idxs[:-1]):
2506 if step is not None:
2507 if i - prev == step:
2508 continue
2509 yield _genslice(start, prev, step)
2510 step = None
2511 continue
2512 if i - prev in [-1, 1]:
2513 step = i - prev
2514 start = prev
2515 continue
2516 else:
2517 yield 's[%d]' % prev
2518 if step is None:
2519 yield 's[%d]' % i
2520 else:
2521 yield _genslice(start, i, step)
2522
2523 test_string = ''.join(map(compat_chr, range(len(example_sig))))
2524 cache_res = func(test_string)
2525 cache_spec = [ord(c) for c in cache_res]
2526 expr_code = ' + '.join(gen_sig_code(cache_spec))
2527 signature_id_tuple = '(%s)' % (
2528 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
2529 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2530 ' return %s\n') % (signature_id_tuple, expr_code)
2531 self.to_screen('Extracted signature function:\n' + code)
2532
2533 def _parse_sig_js(self, jscode):
2534 funcname = self._search_regex(
2535 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2536 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2537 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2538 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2539 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)',
2540 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2541 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2542 # Obsolete patterns
2543 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2544 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2545 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2546 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2547 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2548 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2549 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2550 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2551 jscode, 'Initial JS player signature function name', group='sig')
2552
2553 jsi = JSInterpreter(jscode)
2554 initial_function = jsi.extract_function(funcname)
2555 return lambda s: initial_function([s])
2556
2557 def _decrypt_signature(self, s, video_id, player_url):
2558 """Turn the encrypted s field into a working signature"""
2559 try:
2560 player_id = (player_url, self._signature_cache_id(s))
2561 if player_id not in self._player_cache:
2562 func = self._extract_signature_function(video_id, player_url, s)
2563 self._player_cache[player_id] = func
2564 func = self._player_cache[player_id]
2565 self._print_sig_code(func, s)
2566 return func(s)
2567 except Exception as e:
2568 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2569
2570 def _decrypt_nsig(self, s, video_id, player_url):
2571 """Turn the encrypted n field into a working signature"""
2572 if player_url is None:
2573 raise ExtractorError('Cannot decrypt nsig without player_url')
2574 player_url = urljoin('https://www.youtube.com', player_url)
2575
2576 sig_id = ('nsig_value', s)
2577 if sig_id in self._player_cache:
2578 return self._player_cache[sig_id]
2579
2580 try:
2581 player_id = ('nsig', player_url)
2582 if player_id not in self._player_cache:
2583 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
2584 func = self._player_cache[player_id]
2585 self._player_cache[sig_id] = func(s)
2586 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
2587 return self._player_cache[sig_id]
2588 except Exception as e:
2589 raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
2590
2591 def _extract_n_function_name(self, jscode):
2592 nfunc, idx = self._search_regex(
2593 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2594 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2595 if not idx:
2596 return nfunc
2597 return json.loads(js_to_json(self._search_regex(
2598 rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
2599 f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
2600
2601 def _extract_n_function(self, video_id, player_url):
2602 player_id = self._extract_player_info(player_url)
2603 func_code = self._downloader.cache.load('youtube-nsig', player_id)
2604
2605 if func_code:
2606 jsi = JSInterpreter(func_code)
2607 else:
2608 jscode = self._load_player(video_id, player_url)
2609 funcname = self._extract_n_function_name(jscode)
2610 jsi = JSInterpreter(jscode)
2611 func_code = jsi.extract_function_code(funcname)
2612 self._downloader.cache.store('youtube-nsig', player_id, func_code)
2613
2614 if self.get_param('youtube_print_sig_code'):
2615 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2616
2617 return lambda s: jsi.extract_function_from_code(*func_code)([s])
2618
2619 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2620 """
2621 Extract signatureTimestamp (sts)
2622 Required to tell API what sig/player version is in use.
2623 """
2624 sts = None
2625 if isinstance(ytcfg, dict):
2626 sts = int_or_none(ytcfg.get('STS'))
2627
2628 if not sts:
2629 # Attempt to extract from player
2630 if player_url is None:
2631 error_msg = 'Cannot extract signature timestamp without player_url.'
2632 if fatal:
2633 raise ExtractorError(error_msg)
2634 self.report_warning(error_msg)
2635 return
2636 code = self._load_player(video_id, player_url, fatal=fatal)
2637 if code:
2638 sts = int_or_none(self._search_regex(
2639 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
2640 'JS player signature timestamp', group='sts', fatal=fatal))
2641 return sts
2642
2643 def _mark_watched(self, video_id, player_responses):
2644 playback_url = get_first(
2645 player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
2646 expected_type=url_or_none)
2647 if not playback_url:
2648 self.report_warning('Unable to mark watched')
2649 return
2650 parsed_playback_url = compat_urlparse.urlparse(playback_url)
2651 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
2652
2653 # cpn generation algorithm is reverse engineered from base.js.
2654 # In fact it works even with dummy cpn.
2655 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
2656 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
2657
2658 qs.update({
2659 'ver': ['2'],
2660 'cpn': [cpn],
2661 })
2662 playback_url = compat_urlparse.urlunparse(
2663 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
2664
2665 self._download_webpage(
2666 playback_url, video_id, 'Marking watched',
2667 'Unable to mark watched', fatal=False)
2668
2669 @staticmethod
2670 def _extract_urls(webpage):
2671 # Embedded YouTube player
2672 entries = [
2673 unescapeHTML(mobj.group('url'))
2674 for mobj in re.finditer(r'''(?x)
2675 (?:
2676 <iframe[^>]+?src=|
2677 data-video-url=|
2678 <embed[^>]+?src=|
2679 embedSWF\(?:\s*|
2680 <object[^>]+data=|
2681 new\s+SWFObject\(
2682 )
2683 (["\'])
2684 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
2685 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
2686 \1''', webpage)]
2687
2688 # lazyYT YouTube embed
2689 entries.extend(list(map(
2690 unescapeHTML,
2691 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
2692
2693 # Wordpress "YouTube Video Importer" plugin
2694 matches = re.findall(r'''(?x)<div[^>]+
2695 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
2696 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
2697 entries.extend(m[-1] for m in matches)
2698
2699 return entries
2700
2701 @staticmethod
2702 def _extract_url(webpage):
2703 urls = YoutubeIE._extract_urls(webpage)
2704 return urls[0] if urls else None
2705
2706 @classmethod
2707 def extract_id(cls, url):
2708 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
2709 if mobj is None:
2710 raise ExtractorError('Invalid URL: %s' % url)
2711 return mobj.group('id')
2712
2713 def _extract_chapters_from_json(self, data, duration):
2714 chapter_list = traverse_obj(
2715 data, (
2716 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
2717 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
2718 ), expected_type=list)
2719
2720 return self._extract_chapters(
2721 chapter_list,
2722 chapter_time=lambda chapter: float_or_none(
2723 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
2724 chapter_title=lambda chapter: traverse_obj(
2725 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
2726 duration=duration)
2727
2728 def _extract_chapters_from_engagement_panel(self, data, duration):
2729 content_list = traverse_obj(
2730 data,
2731 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2732 expected_type=list, default=[])
2733 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2734 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2735
2736 return next((
2737 filter(None, (
2738 self._extract_chapters(
2739 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2740 chapter_time, chapter_title, duration)
2741 for contents in content_list
2742 ))), [])
2743
2744 @staticmethod
2745 def _extract_chapters_from_description(description, duration):
2746 chapters = [{'start_time': 0}]
2747 for timestamp, title in re.findall(
2748 r'(?m)^((?:\d+:)?\d{1,2}:\d{2})\b\W*\s(.+?)\s*$', description or ''):
2749 start = parse_duration(timestamp)
2750 if start and title and chapters[-1]['start_time'] < start < duration:
2751 chapters[-1]['end_time'] = start
2752 chapters.append({
2753 'start_time': start,
2754 'title': title,
2755 })
2756 chapters[-1]['end_time'] = duration
2757 return chapters[1:]
2758
2759 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2760 chapters = []
2761 last_chapter = {'start_time': 0}
2762 for idx, chapter in enumerate(chapter_list or []):
2763 title = chapter_title(chapter)
2764 start_time = chapter_time(chapter)
2765 if start_time is None:
2766 continue
2767 last_chapter['end_time'] = start_time
2768 if start_time < last_chapter['start_time']:
2769 if idx == 1:
2770 chapters.pop()
2771 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2772 else:
2773 self.report_warning(f'Invalid start time for chapter "{title}"')
2774 continue
2775 last_chapter = {'start_time': start_time, 'title': title}
2776 chapters.append(last_chapter)
2777 last_chapter['end_time'] = duration
2778 return chapters
2779
2780 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2781 return self._parse_json(self._search_regex(
2782 (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
2783 regex), webpage, name, default='{}'), video_id, fatal=False, lenient=True)
2784
2785 def _extract_comment(self, comment_renderer, parent=None):
2786 comment_id = comment_renderer.get('commentId')
2787 if not comment_id:
2788 return
2789
2790 text = self._get_text(comment_renderer, 'contentText')
2791
2792 # note: timestamp is an estimate calculated from the current time and time_text
2793 timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
2794 author = self._get_text(comment_renderer, 'authorText')
2795 author_id = try_get(comment_renderer,
2796 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2797
2798 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2799 lambda x: x['likeCount']), compat_str)) or 0
2800 author_thumbnail = try_get(comment_renderer,
2801 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2802
2803 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2804 is_favorited = 'creatorHeart' in (try_get(
2805 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2806 return {
2807 'id': comment_id,
2808 'text': text,
2809 'timestamp': timestamp,
2810 'time_text': time_text,
2811 'like_count': votes,
2812 'is_favorited': is_favorited,
2813 'author': author,
2814 'author_id': author_id,
2815 'author_thumbnail': author_thumbnail,
2816 'author_is_uploader': author_is_uploader,
2817 'parent': parent or 'root'
2818 }
2819
2820 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
2821
2822 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
2823
2824 def extract_header(contents):
2825 _continuation = None
2826 for content in contents:
2827 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
2828 expected_comment_count = self._get_count(
2829 comments_header_renderer, 'countText', 'commentsCount')
2830
2831 if expected_comment_count:
2832 tracker['est_total'] = expected_comment_count
2833 self.to_screen(f'Downloading ~{expected_comment_count} comments')
2834 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
2835
2836 sort_menu_item = try_get(
2837 comments_header_renderer,
2838 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2839 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2840
2841 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2842 if not _continuation:
2843 continue
2844
2845 sort_text = str_or_none(sort_menu_item.get('title'))
2846 if not sort_text:
2847 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2848 self.to_screen('Sorting comments by %s' % sort_text.lower())
2849 break
2850 return _continuation
2851
2852 def extract_thread(contents):
2853 if not parent:
2854 tracker['current_page_thread'] = 0
2855 for content in contents:
2856 if not parent and tracker['total_parent_comments'] >= max_parents:
2857 yield
2858 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2859 comment_renderer = get_first(
2860 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
2861 expected_type=dict, default={})
2862
2863 comment = self._extract_comment(comment_renderer, parent)
2864 if not comment:
2865 continue
2866
2867 tracker['running_total'] += 1
2868 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
2869 yield comment
2870
2871 # Attempt to get the replies
2872 comment_replies_renderer = try_get(
2873 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2874
2875 if comment_replies_renderer:
2876 tracker['current_page_thread'] += 1
2877 comment_entries_iter = self._comment_entries(
2878 comment_replies_renderer, ytcfg, video_id,
2879 parent=comment.get('id'), tracker=tracker)
2880 yield from itertools.islice(comment_entries_iter, min(
2881 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
2882
2883 # Keeps track of counts across recursive calls
2884 if not tracker:
2885 tracker = dict(
2886 running_total=0,
2887 est_total=0,
2888 current_page_thread=0,
2889 total_parent_comments=0,
2890 total_reply_comments=0)
2891
2892 # TODO: Deprecated
2893 # YouTube comments have a max depth of 2
2894 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
2895 if max_depth:
2896 self._downloader.deprecation_warning(
2897 '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.')
2898 if max_depth == 1 and parent:
2899 return
2900
2901 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
2902 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
2903
2904 continuation = self._extract_continuation(root_continuation_data)
2905
2906 response = None
2907 is_forced_continuation = False
2908 is_first_continuation = parent is None
2909 if is_first_continuation and not continuation:
2910 # Sometimes you can get comments by generating the continuation yourself,
2911 # even if YouTube initially reports them being disabled - e.g. stories comments.
2912 # Note: if the comment section is actually disabled, YouTube may return a response with
2913 # required check_get_keys missing. So we will disable that check initially in this case.
2914 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
2915 is_forced_continuation = True
2916
2917 for page_num in itertools.count(0):
2918 if not continuation:
2919 break
2920 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
2921 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
2922 if page_num == 0:
2923 if is_first_continuation:
2924 note_prefix = 'Downloading comment section API JSON'
2925 else:
2926 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2927 tracker['current_page_thread'], comment_prog_str)
2928 else:
2929 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2930 ' ' if parent else '', ' replies' if parent else '',
2931 page_num, comment_prog_str)
2932
2933 response = self._extract_response(
2934 item_id=None, query=continuation,
2935 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2936 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
2937 is_forced_continuation = False
2938 continuation_contents = traverse_obj(
2939 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
2940
2941 continuation = None
2942 for continuation_section in continuation_contents:
2943 continuation_items = traverse_obj(
2944 continuation_section,
2945 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
2946 get_all=False, expected_type=list) or []
2947 if is_first_continuation:
2948 continuation = extract_header(continuation_items)
2949 is_first_continuation = False
2950 if continuation:
2951 break
2952 continue
2953
2954 for entry in extract_thread(continuation_items):
2955 if not entry:
2956 return
2957 yield entry
2958 continuation = self._extract_continuation({'contents': continuation_items})
2959 if continuation:
2960 break
2961
2962 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2963 if message and not parent and tracker['running_total'] == 0:
2964 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
2965
2966 @staticmethod
2967 def _generate_comment_continuation(video_id):
2968 """
2969 Generates initial comment section continuation token from given video id
2970 """
2971 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
2972 return base64.b64encode(token.encode()).decode()
2973
2974 def _get_comments(self, ytcfg, video_id, contents, webpage):
2975 """Entry for comment extraction"""
2976 def _real_comment_extract(contents):
2977 renderer = next((
2978 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2979 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2980 yield from self._comment_entries(renderer, ytcfg, video_id)
2981
2982 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2983 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2984
2985 @staticmethod
2986 def _get_checkok_params():
2987 return {'contentCheckOk': True, 'racyCheckOk': True}
2988
2989 @classmethod
2990 def _generate_player_context(cls, sts=None):
2991 context = {
2992 'html5Preference': 'HTML5_PREF_WANTS',
2993 }
2994 if sts is not None:
2995 context['signatureTimestamp'] = sts
2996 return {
2997 'playbackContext': {
2998 'contentPlaybackContext': context
2999 },
3000 **cls._get_checkok_params()
3001 }
3002
3003 @staticmethod
3004 def _is_agegated(player_response):
3005 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3006 return True
3007
3008 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
3009 AGE_GATE_REASONS = (
3010 'confirm your age', 'age-restricted', 'inappropriate', # reason
3011 'age_verification_required', 'age_check_required', # status
3012 )
3013 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3014
3015 @staticmethod
3016 def _is_unplayable(player_response):
3017 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3018
3019 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
3020
3021 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3022 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3023 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3024 headers = self.generate_api_headers(
3025 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3026
3027 yt_query = {
3028 'videoId': video_id,
3029 'params': '8AEB' # enable stories
3030 }
3031 yt_query.update(self._generate_player_context(sts))
3032 return self._extract_response(
3033 item_id=video_id, ep='player', query=yt_query,
3034 ytcfg=player_ytcfg, headers=headers, fatal=True,
3035 default_client=client,
3036 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3037 ) or None
3038
3039 def _get_requested_clients(self, url, smuggled_data):
3040 requested_clients = []
3041 default = ['android', 'web']
3042 allowed_clients = sorted(
3043 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3044 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3045 for client in self._configuration_arg('player_client'):
3046 if client in allowed_clients:
3047 requested_clients.append(client)
3048 elif client == 'default':
3049 requested_clients.extend(default)
3050 elif client == 'all':
3051 requested_clients.extend(allowed_clients)
3052 else:
3053 self.report_warning(f'Skipping unsupported client {client}')
3054 if not requested_clients:
3055 requested_clients = default
3056
3057 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3058 requested_clients.extend(
3059 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3060
3061 return orderedSet(requested_clients)
3062
3063 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
3064 initial_pr = None
3065 if webpage:
3066 initial_pr = self._extract_yt_initial_variable(
3067 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
3068 video_id, 'initial player response')
3069
3070 all_clients = set(clients)
3071 clients = clients[::-1]
3072 prs = []
3073
3074 def append_client(*client_names):
3075 """ Append the first client name that exists but not already used """
3076 for client_name in client_names:
3077 actual_client = _split_innertube_client(client_name)[0]
3078 if actual_client in INNERTUBE_CLIENTS:
3079 if actual_client not in all_clients:
3080 clients.append(client_name)
3081 all_clients.add(actual_client)
3082 return
3083
3084 # Android player_response does not have microFormats which are needed for
3085 # extraction of some data. So we return the initial_pr with formats
3086 # stripped out even if not requested by the user
3087 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3088 if initial_pr:
3089 pr = dict(initial_pr)
3090 pr['streamingData'] = None
3091 prs.append(pr)
3092
3093 last_error = None
3094 tried_iframe_fallback = False
3095 player_url = None
3096 while clients:
3097 client, base_client, variant = _split_innertube_client(clients.pop())
3098 player_ytcfg = master_ytcfg if client == 'web' else {}
3099 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3100 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3101
3102 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3103 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3104 if 'js' in self._configuration_arg('player_skip'):
3105 require_js_player = False
3106 player_url = None
3107
3108 if not player_url and not tried_iframe_fallback and require_js_player:
3109 player_url = self._download_player_url(video_id)
3110 tried_iframe_fallback = True
3111
3112 try:
3113 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3114 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
3115 except ExtractorError as e:
3116 if last_error:
3117 self.report_warning(last_error)
3118 last_error = e
3119 continue
3120
3121 if pr:
3122 prs.append(pr)
3123
3124 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3125 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3126 append_client(f'{base_client}_creator')
3127 elif self._is_agegated(pr):
3128 if variant == 'tv_embedded':
3129 append_client(f'{base_client}_embedded')
3130 elif not variant:
3131 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3132
3133 if last_error:
3134 if not len(prs):
3135 raise last_error
3136 self.report_warning(last_error)
3137 return prs, player_url
3138
3139 def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
3140 itags, stream_ids = {}, []
3141 itag_qualities, res_qualities = {}, {}
3142 q = qualities([
3143 # Normally tiny is the smallest video-only formats. But
3144 # audio-only formats with unknown quality may get tagged as tiny
3145 'tiny',
3146 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3147 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3148 ])
3149 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3150
3151 for fmt in streaming_formats:
3152 if fmt.get('targetDurationSec'):
3153 continue
3154
3155 itag = str_or_none(fmt.get('itag'))
3156 audio_track = fmt.get('audioTrack') or {}
3157 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3158 if stream_id in stream_ids:
3159 continue
3160
3161 quality = fmt.get('quality')
3162 height = int_or_none(fmt.get('height'))
3163 if quality == 'tiny' or not quality:
3164 quality = fmt.get('audioQuality', '').lower() or quality
3165 # The 3gp format (17) in android client has a quality of "small",
3166 # but is actually worse than other formats
3167 if itag == '17':
3168 quality = 'tiny'
3169 if quality:
3170 if itag:
3171 itag_qualities[itag] = quality
3172 if height:
3173 res_qualities[height] = quality
3174 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3175 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3176 # number of fragment that would subsequently requested with (`&sq=N`)
3177 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3178 continue
3179
3180 fmt_url = fmt.get('url')
3181 if not fmt_url:
3182 sc = compat_parse_qs(fmt.get('signatureCipher'))
3183 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3184 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3185 if not all((sc, fmt_url, player_url, encrypted_sig)):
3186 continue
3187 try:
3188 fmt_url += '&%s=%s' % (
3189 traverse_obj(sc, ('sp', -1)) or 'signature',
3190 self._decrypt_signature(encrypted_sig, video_id, player_url)
3191 )
3192 except ExtractorError as e:
3193 self.report_warning('Signature extraction failed: Some formats may be missing', only_once=True)
3194 self.write_debug(e, only_once=True)
3195 continue
3196
3197 query = parse_qs(fmt_url)
3198 throttled = False
3199 if query.get('n'):
3200 try:
3201 fmt_url = update_url_query(fmt_url, {
3202 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
3203 except ExtractorError as e:
3204 self.report_warning(
3205 'nsig extraction failed: You may experience throttling for some formats\n'
3206 f'n = {query["n"][0]} ; player = {player_url}', only_once=True)
3207 self.write_debug(e, only_once=True)
3208 throttled = True
3209
3210 if itag:
3211 itags[itag] = 'https'
3212 stream_ids.append(stream_id)
3213
3214 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3215 language_preference = (
3216 10 if audio_track.get('audioIsDefault') and 10
3217 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3218 else -1)
3219 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3220 # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3221 # Make sure to avoid false positives with small duration differences.
3222 # Eg: __2ABJjxzNo, ySuUZEjARPY
3223 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3224 if is_damaged:
3225 self.report_warning(
3226 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3227 dct = {
3228 'asr': int_or_none(fmt.get('audioSampleRate')),
3229 'filesize': int_or_none(fmt.get('contentLength')),
3230 'format_id': itag,
3231 'format_note': join_nonempty(
3232 '%s%s' % (audio_track.get('displayName') or '',
3233 ' (default)' if language_preference > 0 else ''),
3234 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3235 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3236 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3237 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3238 'fps': int_or_none(fmt.get('fps')) or None,
3239 'height': height,
3240 'quality': q(quality),
3241 'has_drm': bool(fmt.get('drmFamilies')),
3242 'tbr': tbr,
3243 'url': fmt_url,
3244 'width': int_or_none(fmt.get('width')),
3245 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3246 'desc' if language_preference < -1 else ''),
3247 'language_preference': language_preference,
3248 # Strictly de-prioritize damaged and 3gp formats
3249 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3250 }
3251 mime_mobj = re.match(
3252 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3253 if mime_mobj:
3254 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3255 dct.update(parse_codecs(mime_mobj.group(2)))
3256 no_audio = dct.get('acodec') == 'none'
3257 no_video = dct.get('vcodec') == 'none'
3258 if no_audio:
3259 dct['vbr'] = tbr
3260 if no_video:
3261 dct['abr'] = tbr
3262 if no_audio or no_video:
3263 dct['downloader_options'] = {
3264 # Youtube throttles chunks >~10M
3265 'http_chunk_size': 10485760,
3266 }
3267 if dct.get('ext'):
3268 dct['container'] = dct['ext'] + '_dash'
3269 yield dct
3270
3271 live_from_start = is_live and self.get_param('live_from_start')
3272 skip_manifests = self._configuration_arg('skip')
3273 if not self.get_param('youtube_include_hls_manifest', True):
3274 skip_manifests.append('hls')
3275 if not self.get_param('youtube_include_dash_manifest', True):
3276 skip_manifests.append('dash')
3277 get_dash = 'dash' not in skip_manifests and (
3278 not is_live or live_from_start or self._configuration_arg('include_live_dash'))
3279 get_hls = not live_from_start and 'hls' not in skip_manifests
3280
3281 def process_manifest_format(f, proto, itag):
3282 if itag in itags:
3283 if itags[itag] == proto or f'{itag}-{proto}' in itags:
3284 return False
3285 itag = f'{itag}-{proto}'
3286 if itag:
3287 f['format_id'] = itag
3288 itags[itag] = proto
3289
3290 f['quality'] = next((
3291 q(qdict[val])
3292 for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities))
3293 if val in qdict), -1)
3294 return True
3295
3296 for sd in streaming_data:
3297 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
3298 if hls_manifest_url:
3299 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
3300 if process_manifest_format(f, 'hls', self._search_regex(
3301 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3302 yield f
3303
3304 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
3305 if dash_manifest_url:
3306 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
3307 if process_manifest_format(f, 'dash', f['format_id']):
3308 f['filesize'] = int_or_none(self._search_regex(
3309 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3310 if live_from_start:
3311 f['is_from_start'] = True
3312
3313 yield f
3314
3315 def _extract_storyboard(self, player_responses, duration):
3316 spec = get_first(
3317 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3318 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3319 if not base_url:
3320 return
3321 L = len(spec) - 1
3322 for i, args in enumerate(spec):
3323 args = args.split('#')
3324 counts = list(map(int_or_none, args[:5]))
3325 if len(args) != 8 or not all(counts):
3326 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3327 continue
3328 width, height, frame_count, cols, rows = counts
3329 N, sigh = args[6:]
3330
3331 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3332 fragment_count = frame_count / (cols * rows)
3333 fragment_duration = duration / fragment_count
3334 yield {
3335 'format_id': f'sb{i}',
3336 'format_note': 'storyboard',
3337 'ext': 'mhtml',
3338 'protocol': 'mhtml',
3339 'acodec': 'none',
3340 'vcodec': 'none',
3341 'url': url,
3342 'width': width,
3343 'height': height,
3344 'fragments': [{
3345 'url': url.replace('$M', str(j)),
3346 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3347 } for j in range(math.ceil(fragment_count))],
3348 }
3349
3350 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3351 webpage = None
3352 if 'webpage' not in self._configuration_arg('player_skip'):
3353 webpage = self._download_webpage(
3354 webpage_url + '&bpctr=9999999999&has_verified=1&pp=8AEB', video_id, fatal=False)
3355
3356 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3357
3358 player_responses, player_url = self._extract_player_responses(
3359 self._get_requested_clients(url, smuggled_data),
3360 video_id, webpage, master_ytcfg)
3361
3362 return webpage, master_ytcfg, player_responses, player_url
3363
3364 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3365 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3366 is_live = get_first(video_details, 'isLive')
3367 if is_live is None:
3368 is_live = get_first(live_broadcast_details, 'isLiveNow')
3369
3370 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3371 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
3372
3373 return live_broadcast_details, is_live, streaming_data, formats
3374
3375 def _real_extract(self, url):
3376 url, smuggled_data = unsmuggle_url(url, {})
3377 video_id = self._match_id(url)
3378
3379 base_url = self.http_scheme() + '//www.youtube.com/'
3380 webpage_url = base_url + 'watch?v=' + video_id
3381
3382 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3383
3384 playability_statuses = traverse_obj(
3385 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3386
3387 trailer_video_id = get_first(
3388 playability_statuses,
3389 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3390 expected_type=str)
3391 if trailer_video_id:
3392 return self.url_result(
3393 trailer_video_id, self.ie_key(), trailer_video_id)
3394
3395 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3396 if webpage else (lambda x: None))
3397
3398 video_details = traverse_obj(
3399 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3400 microformats = traverse_obj(
3401 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3402 expected_type=dict, default=[])
3403 video_title = (
3404 get_first(video_details, 'title')
3405 or self._get_text(microformats, (..., 'title'))
3406 or search_meta(['og:title', 'twitter:title', 'title']))
3407 video_description = get_first(video_details, 'shortDescription')
3408
3409 multifeed_metadata_list = get_first(
3410 player_responses,
3411 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3412 expected_type=str)
3413 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3414 if self.get_param('noplaylist'):
3415 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3416 else:
3417 entries = []
3418 feed_ids = []
3419 for feed in multifeed_metadata_list.split(','):
3420 # Unquote should take place before split on comma (,) since textual
3421 # fields may contain comma as well (see
3422 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3423 feed_data = compat_parse_qs(
3424 compat_urllib_parse_unquote_plus(feed))
3425
3426 def feed_entry(name):
3427 return try_get(
3428 feed_data, lambda x: x[name][0], compat_str)
3429
3430 feed_id = feed_entry('id')
3431 if not feed_id:
3432 continue
3433 feed_title = feed_entry('title')
3434 title = video_title
3435 if feed_title:
3436 title += ' (%s)' % feed_title
3437 entries.append({
3438 '_type': 'url_transparent',
3439 'ie_key': 'Youtube',
3440 'url': smuggle_url(
3441 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3442 {'force_singlefeed': True}),
3443 'title': title,
3444 })
3445 feed_ids.append(feed_id)
3446 self.to_screen(
3447 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3448 % (', '.join(feed_ids), video_id))
3449 return self.playlist_result(
3450 entries, video_id, video_title, video_description)
3451
3452 duration = int_or_none(
3453 get_first(video_details, 'lengthSeconds')
3454 or get_first(microformats, 'lengthSeconds')
3455 or parse_duration(search_meta('duration'))) or None
3456
3457 if get_first(video_details, 'isPostLiveDvr'):
3458 self.write_debug('Video is in Post-Live Manifestless mode')
3459 if duration or 0 > 4 * 3600:
3460 self.report_warning(
3461 'The livestream has not finished processing. Only 4 hours of the video can be currently downloaded. '
3462 'This is a known issue and patches are welcome')
3463
3464 live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
3465 video_id, microformats, video_details, player_responses, player_url, duration)
3466
3467 if not formats:
3468 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3469 self.report_drm(video_id)
3470 pemr = get_first(
3471 playability_statuses,
3472 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3473 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3474 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3475 if subreason:
3476 if subreason == 'The uploader has not made this video available in your country.':
3477 countries = get_first(microformats, 'availableCountries')
3478 if not countries:
3479 regions_allowed = search_meta('regionsAllowed')
3480 countries = regions_allowed.split(',') if regions_allowed else None
3481 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3482 reason += f'. {subreason}'
3483 if reason:
3484 self.raise_no_formats(reason, expected=True)
3485
3486 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3487 if not keywords and webpage:
3488 keywords = [
3489 unescapeHTML(m.group('content'))
3490 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3491 for keyword in keywords:
3492 if keyword.startswith('yt:stretch='):
3493 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3494 if mobj:
3495 # NB: float is intentional for forcing float division
3496 w, h = (float(v) for v in mobj.groups())
3497 if w > 0 and h > 0:
3498 ratio = w / h
3499 for f in formats:
3500 if f.get('vcodec') != 'none':
3501 f['stretched_ratio'] = ratio
3502 break
3503 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3504 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3505 if thumbnail_url:
3506 thumbnails.append({
3507 'url': thumbnail_url,
3508 })
3509 original_thumbnails = thumbnails.copy()
3510
3511 # The best resolution thumbnails sometimes does not appear in the webpage
3512 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3513 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3514 thumbnail_names = [
3515 # While the *1,*2,*3 thumbnails are just below their correspnding "*default" variants
3516 # in resolution, these are not the custom thumbnail. So de-prioritize them
3517 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3518 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3519 ]
3520 n_thumbnail_names = len(thumbnail_names)
3521 thumbnails.extend({
3522 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3523 video_id=video_id, name=name, ext=ext,
3524 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
3525 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3526 for thumb in thumbnails:
3527 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3528 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3529 self._remove_duplicate_formats(thumbnails)
3530 self._downloader._sort_thumbnails(original_thumbnails)
3531
3532 category = get_first(microformats, 'category') or search_meta('genre')
3533 channel_id = str_or_none(
3534 get_first(video_details, 'channelId')
3535 or get_first(microformats, 'externalChannelId')
3536 or search_meta('channelId'))
3537 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3538
3539 live_content = get_first(video_details, 'isLiveContent')
3540 is_upcoming = get_first(video_details, 'isUpcoming')
3541 if is_live is None:
3542 if is_upcoming or live_content is False:
3543 is_live = False
3544 if is_upcoming is None and (live_content or is_live):
3545 is_upcoming = False
3546 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3547 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3548 if not duration and live_end_time and live_start_time:
3549 duration = live_end_time - live_start_time
3550
3551 if is_live and self.get_param('live_from_start'):
3552 self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
3553
3554 formats.extend(self._extract_storyboard(player_responses, duration))
3555
3556 # Source is given priority since formats that throttle are given lower source_preference
3557 # When throttling issue is fully fixed, remove this
3558 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
3559
3560 info = {
3561 'id': video_id,
3562 'title': video_title,
3563 'formats': formats,
3564 'thumbnails': thumbnails,
3565 # The best thumbnail that we are sure exists. Prevents unnecessary
3566 # URL checking if user don't care about getting the best possible thumbnail
3567 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
3568 'description': video_description,
3569 'uploader': get_first(video_details, 'author'),
3570 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
3571 'uploader_url': owner_profile_url,
3572 'channel_id': channel_id,
3573 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
3574 'duration': duration,
3575 'view_count': int_or_none(
3576 get_first((video_details, microformats), (..., 'viewCount'))
3577 or search_meta('interactionCount')),
3578 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
3579 'age_limit': 18 if (
3580 get_first(microformats, 'isFamilySafe') is False
3581 or search_meta('isFamilyFriendly') == 'false'
3582 or search_meta('og:restrictions:age') == '18+') else 0,
3583 'webpage_url': webpage_url,
3584 'categories': [category] if category else None,
3585 'tags': keywords,
3586 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
3587 'is_live': is_live,
3588 'was_live': (False if is_live or is_upcoming or live_content is False
3589 else None if is_live is None or is_upcoming is None
3590 else live_content),
3591 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
3592 'release_timestamp': live_start_time,
3593 }
3594
3595 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
3596 if pctr:
3597 def get_lang_code(track):
3598 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
3599 or track.get('languageCode'))
3600
3601 # Converted into dicts to remove duplicates
3602 captions = {
3603 get_lang_code(sub): sub
3604 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
3605 translation_languages = {
3606 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
3607 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
3608
3609 def process_language(container, base_url, lang_code, sub_name, query):
3610 lang_subs = container.setdefault(lang_code, [])
3611 for fmt in self._SUBTITLE_FORMATS:
3612 query.update({
3613 'fmt': fmt,
3614 })
3615 lang_subs.append({
3616 'ext': fmt,
3617 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
3618 'name': sub_name,
3619 })
3620
3621 subtitles, automatic_captions = {}, {}
3622 for lang_code, caption_track in captions.items():
3623 base_url = caption_track.get('baseUrl')
3624 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
3625 if not base_url:
3626 continue
3627 lang_name = self._get_text(caption_track, 'name', max_runs=1)
3628 if caption_track.get('kind') != 'asr':
3629 if not lang_code:
3630 continue
3631 process_language(
3632 subtitles, base_url, lang_code, lang_name, {})
3633 if not caption_track.get('isTranslatable'):
3634 continue
3635 for trans_code, trans_name in translation_languages.items():
3636 if not trans_code:
3637 continue
3638 orig_trans_code = trans_code
3639 if caption_track.get('kind') != 'asr':
3640 if 'translated_subs' in self._configuration_arg('skip'):
3641 continue
3642 trans_code += f'-{lang_code}'
3643 trans_name += format_field(lang_name, template=' from %s')
3644 # Add an "-orig" label to the original language so that it can be distinguished.
3645 # The subs are returned without "-orig" as well for compatibility
3646 if lang_code == f'a-{orig_trans_code}':
3647 process_language(
3648 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
3649 # Setting tlang=lang returns damaged subtitles.
3650 process_language(automatic_captions, base_url, trans_code, trans_name,
3651 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
3652 info['automatic_captions'] = automatic_captions
3653 info['subtitles'] = subtitles
3654
3655 parsed_url = compat_urllib_parse_urlparse(url)
3656 for component in [parsed_url.fragment, parsed_url.query]:
3657 query = compat_parse_qs(component)
3658 for k, v in query.items():
3659 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
3660 d_k += '_time'
3661 if d_k not in info and k in s_ks:
3662 info[d_k] = parse_duration(query[k][0])
3663
3664 # Youtube Music Auto-generated description
3665 if video_description:
3666 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)
3667 if mobj:
3668 release_year = mobj.group('release_year')
3669 release_date = mobj.group('release_date')
3670 if release_date:
3671 release_date = release_date.replace('-', '')
3672 if not release_year:
3673 release_year = release_date[:4]
3674 info.update({
3675 'album': mobj.group('album'.strip()),
3676 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
3677 'track': mobj.group('track').strip(),
3678 'release_date': release_date,
3679 'release_year': int_or_none(release_year),
3680 })
3681
3682 initial_data = None
3683 if webpage:
3684 initial_data = self._extract_yt_initial_variable(
3685 webpage, self._YT_INITIAL_DATA_RE, video_id,
3686 'yt initial data')
3687 if not initial_data:
3688 query = {'videoId': video_id}
3689 query.update(self._get_checkok_params())
3690 initial_data = self._extract_response(
3691 item_id=video_id, ep='next', fatal=False,
3692 ytcfg=master_ytcfg, query=query,
3693 headers=self.generate_api_headers(ytcfg=master_ytcfg),
3694 note='Downloading initial data API JSON')
3695
3696 try: # This will error if there is no livechat
3697 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
3698 except (KeyError, IndexError, TypeError):
3699 pass
3700 else:
3701 info.setdefault('subtitles', {})['live_chat'] = [{
3702 'url': f'https://www.youtube.com/watch?v={video_id}', # url is needed to set cookies
3703 'video_id': video_id,
3704 'ext': 'json',
3705 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
3706 }]
3707
3708 if initial_data:
3709 info['chapters'] = (
3710 self._extract_chapters_from_json(initial_data, duration)
3711 or self._extract_chapters_from_engagement_panel(initial_data, duration)
3712 or self._extract_chapters_from_description(video_description, duration)
3713 or None)
3714
3715 contents = traverse_obj(
3716 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
3717 expected_type=list, default=[])
3718
3719 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
3720 if vpir:
3721 stl = vpir.get('superTitleLink')
3722 if stl:
3723 stl = self._get_text(stl)
3724 if try_get(
3725 vpir,
3726 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
3727 info['location'] = stl
3728 else:
3729 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
3730 if mobj:
3731 info.update({
3732 'series': mobj.group(1),
3733 'season_number': int(mobj.group(2)),
3734 'episode_number': int(mobj.group(3)),
3735 })
3736 for tlb in (try_get(
3737 vpir,
3738 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
3739 list) or []):
3740 tbr = tlb.get('toggleButtonRenderer') or {}
3741 for getter, regex in [(
3742 lambda x: x['defaultText']['accessibility']['accessibilityData'],
3743 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
3744 lambda x: x['accessibility'],
3745 lambda x: x['accessibilityData']['accessibilityData'],
3746 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
3747 label = (try_get(tbr, getter, dict) or {}).get('label')
3748 if label:
3749 mobj = re.match(regex, label)
3750 if mobj:
3751 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
3752 break
3753 sbr_tooltip = try_get(
3754 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
3755 if sbr_tooltip:
3756 like_count, dislike_count = sbr_tooltip.split(' / ')
3757 info.update({
3758 'like_count': str_to_int(like_count),
3759 'dislike_count': str_to_int(dislike_count),
3760 })
3761 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
3762 if vsir:
3763 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
3764 info.update({
3765 'channel': self._get_text(vor, 'title'),
3766 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
3767
3768 rows = try_get(
3769 vsir,
3770 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
3771 list) or []
3772 multiple_songs = False
3773 for row in rows:
3774 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
3775 multiple_songs = True
3776 break
3777 for row in rows:
3778 mrr = row.get('metadataRowRenderer') or {}
3779 mrr_title = mrr.get('title')
3780 if not mrr_title:
3781 continue
3782 mrr_title = self._get_text(mrr, 'title')
3783 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3784 if mrr_title == 'License':
3785 info['license'] = mrr_contents_text
3786 elif not multiple_songs:
3787 if mrr_title == 'Album':
3788 info['album'] = mrr_contents_text
3789 elif mrr_title == 'Artist':
3790 info['artist'] = mrr_contents_text
3791 elif mrr_title == 'Song':
3792 info['track'] = mrr_contents_text
3793
3794 fallbacks = {
3795 'channel': 'uploader',
3796 'channel_id': 'uploader_id',
3797 'channel_url': 'uploader_url',
3798 }
3799
3800 # The upload date for scheduled, live and past live streams / premieres in microformats
3801 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
3802 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
3803 upload_date = (
3804 unified_strdate(get_first(microformats, 'uploadDate'))
3805 or unified_strdate(search_meta('uploadDate')))
3806 if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
3807 upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') or upload_date
3808 info['upload_date'] = upload_date
3809
3810 for to, frm in fallbacks.items():
3811 if not info.get(to):
3812 info[to] = info.get(frm)
3813
3814 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3815 v = info.get(s_k)
3816 if v:
3817 info[d_k] = v
3818
3819 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3820 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3821 is_membersonly = None
3822 is_premium = None
3823 if initial_data and is_private is not None:
3824 is_membersonly = False
3825 is_premium = False
3826 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3827 badge_labels = set()
3828 for content in contents:
3829 if not isinstance(content, dict):
3830 continue
3831 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3832 for badge_label in badge_labels:
3833 if badge_label.lower() == 'members only':
3834 is_membersonly = True
3835 elif badge_label.lower() == 'premium':
3836 is_premium = True
3837 elif badge_label.lower() == 'unlisted':
3838 is_unlisted = True
3839
3840 info['availability'] = self._availability(
3841 is_private=is_private,
3842 needs_premium=is_premium,
3843 needs_subscription=is_membersonly,
3844 needs_auth=info['age_limit'] >= 18,
3845 is_unlisted=None if is_private is None else is_unlisted)
3846
3847 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3848
3849 self.mark_watched(video_id, player_responses)
3850
3851 return info
3852
3853
3854class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
3855
3856 @staticmethod
3857 def passthrough_smuggled_data(func):
3858 def _smuggle(entries, smuggled_data):
3859 for entry in entries:
3860 # TODO: Convert URL to music.youtube instead.
3861 # Do we need to passthrough any other smuggled_data?
3862 entry['url'] = smuggle_url(entry['url'], smuggled_data)
3863 yield entry
3864
3865 @functools.wraps(func)
3866 def wrapper(self, url):
3867 url, smuggled_data = unsmuggle_url(url, {})
3868 if self.is_music_url(url):
3869 smuggled_data['is_music_url'] = True
3870 info_dict = func(self, url, smuggled_data)
3871 if smuggled_data and info_dict.get('entries'):
3872 info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
3873 return info_dict
3874 return wrapper
3875
3876 def _extract_channel_id(self, webpage):
3877 channel_id = self._html_search_meta(
3878 'channelId', webpage, 'channel id', default=None)
3879 if channel_id:
3880 return channel_id
3881 channel_url = self._html_search_meta(
3882 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3883 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3884 'twitter:app:url:googleplay'), webpage, 'channel url')
3885 return self._search_regex(
3886 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3887 channel_url, 'channel id')
3888
3889 @staticmethod
3890 def _extract_basic_item_renderer(item):
3891 # Modified from _extract_grid_item_renderer
3892 known_basic_renderers = (
3893 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
3894 )
3895 for key, renderer in item.items():
3896 if not isinstance(renderer, dict):
3897 continue
3898 elif key in known_basic_renderers:
3899 return renderer
3900 elif key.startswith('grid') and key.endswith('Renderer'):
3901 return renderer
3902
3903 def _grid_entries(self, grid_renderer):
3904 for item in grid_renderer['items']:
3905 if not isinstance(item, dict):
3906 continue
3907 renderer = self._extract_basic_item_renderer(item)
3908 if not isinstance(renderer, dict):
3909 continue
3910 title = self._get_text(renderer, 'title')
3911
3912 # playlist
3913 playlist_id = renderer.get('playlistId')
3914 if playlist_id:
3915 yield self.url_result(
3916 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3917 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3918 video_title=title)
3919 continue
3920 # video
3921 video_id = renderer.get('videoId')
3922 if video_id:
3923 yield self._extract_video(renderer)
3924 continue
3925 # channel
3926 channel_id = renderer.get('channelId')
3927 if channel_id:
3928 yield self.url_result(
3929 'https://www.youtube.com/channel/%s' % channel_id,
3930 ie=YoutubeTabIE.ie_key(), video_title=title)
3931 continue
3932 # generic endpoint URL support
3933 ep_url = urljoin('https://www.youtube.com/', try_get(
3934 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3935 compat_str))
3936 if ep_url:
3937 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3938 if ie.suitable(ep_url):
3939 yield self.url_result(
3940 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3941 break
3942
3943 def _music_reponsive_list_entry(self, renderer):
3944 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
3945 if video_id:
3946 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
3947 ie=YoutubeIE.ie_key(), video_id=video_id)
3948 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
3949 if playlist_id:
3950 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
3951 if video_id:
3952 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
3953 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3954 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
3955 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3956 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
3957 if browse_id:
3958 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
3959 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
3960
3961 def _shelf_entries_from_content(self, shelf_renderer):
3962 content = shelf_renderer.get('content')
3963 if not isinstance(content, dict):
3964 return
3965 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3966 if renderer:
3967 # TODO: add support for nested playlists so each shelf is processed
3968 # as separate playlist
3969 # TODO: this includes only first N items
3970 yield from self._grid_entries(renderer)
3971 renderer = content.get('horizontalListRenderer')
3972 if renderer:
3973 # TODO
3974 pass
3975
3976 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3977 ep = try_get(
3978 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3979 compat_str)
3980 shelf_url = urljoin('https://www.youtube.com', ep)
3981 if shelf_url:
3982 # Skipping links to another channels, note that checking for
3983 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3984 # will not work
3985 if skip_channels and '/channels?' in shelf_url:
3986 return
3987 title = self._get_text(shelf_renderer, 'title')
3988 yield self.url_result(shelf_url, video_title=title)
3989 # Shelf may not contain shelf URL, fallback to extraction from content
3990 yield from self._shelf_entries_from_content(shelf_renderer)
3991
3992 def _playlist_entries(self, video_list_renderer):
3993 for content in video_list_renderer['contents']:
3994 if not isinstance(content, dict):
3995 continue
3996 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3997 if not isinstance(renderer, dict):
3998 continue
3999 video_id = renderer.get('videoId')
4000 if not video_id:
4001 continue
4002 yield self._extract_video(renderer)
4003
4004 def _rich_entries(self, rich_grid_renderer):
4005 renderer = try_get(
4006 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
4007 video_id = renderer.get('videoId')
4008 if not video_id:
4009 return
4010 yield self._extract_video(renderer)
4011
4012 def _video_entry(self, video_renderer):
4013 video_id = video_renderer.get('videoId')
4014 if video_id:
4015 return self._extract_video(video_renderer)
4016
4017 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4018 url = urljoin('https://youtube.com', traverse_obj(
4019 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4020 if url:
4021 return self.url_result(
4022 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4023
4024 def _post_thread_entries(self, post_thread_renderer):
4025 post_renderer = try_get(
4026 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4027 if not post_renderer:
4028 return
4029 # video attachment
4030 video_renderer = try_get(
4031 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4032 video_id = video_renderer.get('videoId')
4033 if video_id:
4034 entry = self._extract_video(video_renderer)
4035 if entry:
4036 yield entry
4037 # playlist attachment
4038 playlist_id = try_get(
4039 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
4040 if playlist_id:
4041 yield self.url_result(
4042 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4043 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4044 # inline video links
4045 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4046 for run in runs:
4047 if not isinstance(run, dict):
4048 continue
4049 ep_url = try_get(
4050 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
4051 if not ep_url:
4052 continue
4053 if not YoutubeIE.suitable(ep_url):
4054 continue
4055 ep_video_id = YoutubeIE._match_id(ep_url)
4056 if video_id == ep_video_id:
4057 continue
4058 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4059
4060 def _post_thread_continuation_entries(self, post_thread_continuation):
4061 contents = post_thread_continuation.get('contents')
4062 if not isinstance(contents, list):
4063 return
4064 for content in contents:
4065 renderer = content.get('backstagePostThreadRenderer')
4066 if not isinstance(renderer, dict):
4067 continue
4068 yield from self._post_thread_entries(renderer)
4069
4070 r''' # unused
4071 def _rich_grid_entries(self, contents):
4072 for content in contents:
4073 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4074 if video_renderer:
4075 entry = self._video_entry(video_renderer)
4076 if entry:
4077 yield entry
4078 '''
4079
4080 def _extract_entries(self, parent_renderer, continuation_list):
4081 # continuation_list is modified in-place with continuation_list = [continuation_token]
4082 continuation_list[:] = [None]
4083 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4084 for content in contents:
4085 if not isinstance(content, dict):
4086 continue
4087 is_renderer = traverse_obj(
4088 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4089 expected_type=dict)
4090 if not is_renderer:
4091 renderer = content.get('richItemRenderer')
4092 if renderer:
4093 for entry in self._rich_entries(renderer):
4094 yield entry
4095 continuation_list[0] = self._extract_continuation(parent_renderer)
4096 continue
4097 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4098 for isr_content in isr_contents:
4099 if not isinstance(isr_content, dict):
4100 continue
4101
4102 known_renderers = {
4103 'playlistVideoListRenderer': self._playlist_entries,
4104 'gridRenderer': self._grid_entries,
4105 'reelShelfRenderer': self._grid_entries,
4106 'shelfRenderer': self._shelf_entries,
4107 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4108 'backstagePostThreadRenderer': self._post_thread_entries,
4109 'videoRenderer': lambda x: [self._video_entry(x)],
4110 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4111 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4112 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4113 }
4114 for key, renderer in isr_content.items():
4115 if key not in known_renderers:
4116 continue
4117 for entry in known_renderers[key](renderer):
4118 if entry:
4119 yield entry
4120 continuation_list[0] = self._extract_continuation(renderer)
4121 break
4122
4123 if not continuation_list[0]:
4124 continuation_list[0] = self._extract_continuation(is_renderer)
4125
4126 if not continuation_list[0]:
4127 continuation_list[0] = self._extract_continuation(parent_renderer)
4128
4129 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4130 continuation_list = [None]
4131 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4132 tab_content = try_get(tab, lambda x: x['content'], dict)
4133 if not tab_content:
4134 return
4135 parent_renderer = (
4136 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4137 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4138 yield from extract_entries(parent_renderer)
4139 continuation = continuation_list[0]
4140
4141 for page_num in itertools.count(1):
4142 if not continuation:
4143 break
4144 headers = self.generate_api_headers(
4145 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4146 response = self._extract_response(
4147 item_id=f'{item_id} page {page_num}',
4148 query=continuation, headers=headers, ytcfg=ytcfg,
4149 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4150
4151 if not response:
4152 break
4153 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4154 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4155 visitor_data = self._extract_visitor_data(response) or visitor_data
4156
4157 known_continuation_renderers = {
4158 'playlistVideoListContinuation': self._playlist_entries,
4159 'gridContinuation': self._grid_entries,
4160 'itemSectionContinuation': self._post_thread_continuation_entries,
4161 'sectionListContinuation': extract_entries, # for feeds
4162 }
4163 continuation_contents = try_get(
4164 response, lambda x: x['continuationContents'], dict) or {}
4165 continuation_renderer = None
4166 for key, value in continuation_contents.items():
4167 if key not in known_continuation_renderers:
4168 continue
4169 continuation_renderer = value
4170 continuation_list = [None]
4171 yield from known_continuation_renderers[key](continuation_renderer)
4172 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
4173 break
4174 if continuation_renderer:
4175 continue
4176
4177 known_renderers = {
4178 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4179 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4180 'gridVideoRenderer': (self._grid_entries, 'items'),
4181 'gridChannelRenderer': (self._grid_entries, 'items'),
4182 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4183 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4184 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4185 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
4186 }
4187 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4188 continuation_items = try_get(
4189 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
4190 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
4191 video_items_renderer = None
4192 for key, value in continuation_item.items():
4193 if key not in known_renderers:
4194 continue
4195 video_items_renderer = {known_renderers[key][1]: continuation_items}
4196 continuation_list = [None]
4197 yield from known_renderers[key][0](video_items_renderer)
4198 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4199 break
4200 if video_items_renderer:
4201 continue
4202 break
4203
4204 @staticmethod
4205 def _extract_selected_tab(tabs, fatal=True):
4206 for tab in tabs:
4207 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
4208 if renderer.get('selected') is True:
4209 return renderer
4210 else:
4211 if fatal:
4212 raise ExtractorError('Unable to find selected tab')
4213
4214 def _extract_uploader(self, data):
4215 uploader = {}
4216 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
4217 owner = try_get(
4218 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
4219 if owner:
4220 owner_text = owner.get('text')
4221 uploader['uploader'] = self._search_regex(
4222 r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
4223 uploader['uploader_id'] = try_get(
4224 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
4225 uploader['uploader_url'] = urljoin(
4226 'https://www.youtube.com/',
4227 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
4228 return {k: v for k, v in uploader.items() if v is not None}
4229
4230 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4231 playlist_id = title = description = channel_url = channel_name = channel_id = None
4232 tags = []
4233
4234 selected_tab = self._extract_selected_tab(tabs)
4235 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4236 renderer = try_get(
4237 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
4238 if renderer:
4239 channel_name = renderer.get('title')
4240 channel_url = renderer.get('channelUrl')
4241 channel_id = renderer.get('externalId')
4242 else:
4243 renderer = try_get(
4244 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
4245
4246 if renderer:
4247 title = renderer.get('title')
4248 description = renderer.get('description', '')
4249 playlist_id = channel_id
4250 tags = renderer.get('keywords', '').split()
4251
4252 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4253 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4254 def _get_uncropped(url):
4255 return url_or_none((url or '').split('=')[0] + '=s0')
4256
4257 avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
4258 if avatar_thumbnails:
4259 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4260 if uncropped_avatar:
4261 avatar_thumbnails.append({
4262 'url': uncropped_avatar,
4263 'id': 'avatar_uncropped',
4264 'preference': 1
4265 })
4266
4267 channel_banners = self._extract_thumbnails(
4268 data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
4269 for banner in channel_banners:
4270 banner['preference'] = -10
4271
4272 if channel_banners:
4273 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4274 if uncropped_banner:
4275 channel_banners.append({
4276 'url': uncropped_banner,
4277 'id': 'banner_uncropped',
4278 'preference': -5
4279 })
4280
4281 primary_thumbnails = self._extract_thumbnails(
4282 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4283
4284 if playlist_id is None:
4285 playlist_id = item_id
4286
4287 playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
4288 last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
4289 if title is None:
4290 title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
4291 title += format_field(selected_tab, 'title', ' - %s')
4292 title += format_field(selected_tab, 'expandedText', ' - %s')
4293
4294 metadata = {
4295 'playlist_id': playlist_id,
4296 'playlist_title': title,
4297 'playlist_description': description,
4298 'uploader': channel_name,
4299 'uploader_id': channel_id,
4300 'uploader_url': channel_url,
4301 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
4302 'tags': tags,
4303 'view_count': self._get_count(playlist_stats, 1),
4304 'availability': self._extract_availability(data),
4305 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
4306 'playlist_count': self._get_count(playlist_stats, 0),
4307 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4308 }
4309 if not channel_id:
4310 metadata.update(self._extract_uploader(data))
4311 metadata.update({
4312 'channel': metadata['uploader'],
4313 'channel_id': metadata['uploader_id'],
4314 'channel_url': metadata['uploader_url']})
4315 return self.playlist_result(
4316 self._entries(
4317 selected_tab, playlist_id, ytcfg,
4318 self._extract_account_syncid(ytcfg, data),
4319 self._extract_visitor_data(data, ytcfg)),
4320 **metadata)
4321
4322 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4323 first_id = last_id = response = None
4324 for page_num in itertools.count(1):
4325 videos = list(self._playlist_entries(playlist))
4326 if not videos:
4327 return
4328 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4329 if start >= len(videos):
4330 return
4331 yield from videos[start:]
4332 first_id = first_id or videos[0]['id']
4333 last_id = videos[-1]['id']
4334 watch_endpoint = try_get(
4335 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4336 headers = self.generate_api_headers(
4337 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4338 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4339 query = {
4340 'playlistId': playlist_id,
4341 'videoId': watch_endpoint.get('videoId') or last_id,
4342 'index': watch_endpoint.get('index') or len(videos),
4343 'params': watch_endpoint.get('params') or 'OAE%3D'
4344 }
4345 response = self._extract_response(
4346 item_id='%s page %d' % (playlist_id, page_num),
4347 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4348 check_get_keys='contents'
4349 )
4350 playlist = try_get(
4351 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4352
4353 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4354 title = playlist.get('title') or try_get(
4355 data, lambda x: x['titleText']['simpleText'], compat_str)
4356 playlist_id = playlist.get('playlistId') or item_id
4357
4358 # Delegating everything except mix playlists to regular tab-based playlist URL
4359 playlist_url = urljoin(url, try_get(
4360 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4361 compat_str))
4362
4363 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4364 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4365 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4366
4367 if playlist_url and playlist_url != url and not is_known_unviewable:
4368 return self.url_result(
4369 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4370 video_title=title)
4371
4372 return self.playlist_result(
4373 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4374 playlist_id=playlist_id, playlist_title=title)
4375
4376 def _extract_availability(self, data):
4377 """
4378 Gets the availability of a given playlist/tab.
4379 Note: Unless YouTube tells us explicitly, we do not assume it is public
4380 @param data: response
4381 """
4382 is_private = is_unlisted = None
4383 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4384 badge_labels = self._extract_badges(renderer)
4385
4386 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4387 privacy_dropdown_entries = try_get(
4388 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4389 for renderer_dict in privacy_dropdown_entries:
4390 is_selected = try_get(
4391 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4392 if not is_selected:
4393 continue
4394 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4395 if label:
4396 badge_labels.add(label.lower())
4397 break
4398
4399 for badge_label in badge_labels:
4400 if badge_label == 'unlisted':
4401 is_unlisted = True
4402 elif badge_label == 'private':
4403 is_private = True
4404 elif badge_label == 'public':
4405 is_unlisted = is_private = False
4406 return self._availability(is_private, False, False, False, is_unlisted)
4407
4408 @staticmethod
4409 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4410 sidebar_renderer = try_get(
4411 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4412 for item in sidebar_renderer:
4413 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4414 if renderer:
4415 return renderer
4416
4417 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4418 """
4419 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4420 """
4421 browse_id = params = None
4422 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4423 if not renderer:
4424 return
4425 menu_renderer = try_get(
4426 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4427 for menu_item in menu_renderer:
4428 if not isinstance(menu_item, dict):
4429 continue
4430 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4431 text = try_get(
4432 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4433 if not text or text.lower() != 'show unavailable videos':
4434 continue
4435 browse_endpoint = try_get(
4436 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4437 browse_id = browse_endpoint.get('browseId')
4438 params = browse_endpoint.get('params')
4439 break
4440
4441 headers = self.generate_api_headers(
4442 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4443 visitor_data=self._extract_visitor_data(data, ytcfg))
4444 query = {
4445 'params': params or 'wgYCCAA=',
4446 'browseId': browse_id or 'VL%s' % item_id
4447 }
4448 return self._extract_response(
4449 item_id=item_id, headers=headers, query=query,
4450 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4451 note='Downloading API JSON with unavailable videos')
4452
4453 @functools.cached_property
4454 def skip_webpage(self):
4455 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4456
4457 def _extract_webpage(self, url, item_id, fatal=True):
4458 retries = self.get_param('extractor_retries', 3)
4459 count = -1
4460 webpage = data = last_error = None
4461 while count < retries:
4462 count += 1
4463 # Sometimes youtube returns a webpage with incomplete ytInitialData
4464 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4465 if last_error:
4466 self.report_warning('%s. Retrying ...' % last_error)
4467 try:
4468 webpage = self._download_webpage(
4469 url, item_id,
4470 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4471 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4472 except ExtractorError as e:
4473 if isinstance(e.cause, network_exceptions):
4474 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4475 last_error = error_to_compat_str(e.cause or e.msg)
4476 if count < retries:
4477 continue
4478 if fatal:
4479 raise
4480 self.report_warning(error_to_compat_str(e))
4481 break
4482 else:
4483 try:
4484 self._extract_and_report_alerts(data)
4485 except ExtractorError as e:
4486 if fatal:
4487 raise
4488 self.report_warning(error_to_compat_str(e))
4489 break
4490
4491 if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
4492 break
4493
4494 last_error = 'Incomplete yt initial data received'
4495 if count >= retries:
4496 if fatal:
4497 raise ExtractorError(last_error)
4498 self.report_warning(last_error)
4499 break
4500
4501 return webpage, data
4502
4503 def _report_playlist_authcheck(self, ytcfg, fatal=True):
4504 """Use if failed to extract ytcfg (and data) from initial webpage"""
4505 if not ytcfg and self.is_authenticated:
4506 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4507 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4508 raise ExtractorError(
4509 f'{msg}. If you are not downloading private content, or '
4510 'your cookies are only for the first account and channel,'
4511 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4512 expected=True)
4513 self.report_warning(msg, only_once=True)
4514
4515 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4516 data = None
4517 if not self.skip_webpage:
4518 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4519 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4520 # Reject webpage data if redirected to home page without explicitly requesting
4521 selected_tab = self._extract_selected_tab(traverse_obj(
4522 data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
4523 if (url != 'https://www.youtube.com/feed/recommended'
4524 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
4525 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4526 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4527 if fatal:
4528 raise ExtractorError(msg, expected=True)
4529 self.report_warning(msg, only_once=True)
4530 if not data:
4531 self._report_playlist_authcheck(ytcfg, fatal=fatal)
4532 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4533 return data, ytcfg
4534
4535 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4536 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4537 resolve_response = self._extract_response(
4538 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4539 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4540 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4541 for ep_key, ep in endpoints.items():
4542 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4543 if params:
4544 return self._extract_response(
4545 item_id=item_id, query=params, ep=ep, headers=headers,
4546 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4547 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
4548 err_note = 'Failed to resolve url (does the playlist exist?)'
4549 if fatal:
4550 raise ExtractorError(err_note, expected=True)
4551 self.report_warning(err_note, item_id)
4552
4553 _SEARCH_PARAMS = None
4554
4555 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
4556 data = {'query': query}
4557 if params is NO_DEFAULT:
4558 params = self._SEARCH_PARAMS
4559 if params:
4560 data['params'] = params
4561
4562 content_keys = (
4563 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
4564 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
4565 # ytmusic search
4566 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
4567 ('continuationContents', ),
4568 )
4569 display_id = f'query "{query}"'
4570 check_get_keys = tuple({keys[0] for keys in content_keys})
4571 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
4572 self._report_playlist_authcheck(ytcfg, fatal=False)
4573
4574 continuation_list = [None]
4575 search = None
4576 for page_num in itertools.count(1):
4577 data.update(continuation_list[0] or {})
4578 headers = self.generate_api_headers(
4579 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
4580 search = self._extract_response(
4581 item_id=f'{display_id} page {page_num}', ep='search', query=data,
4582 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
4583 slr_contents = traverse_obj(search, *content_keys)
4584 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
4585 if not continuation_list[0]:
4586 break
4587
4588
4589class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
4590 IE_DESC = 'YouTube Tabs'
4591 _VALID_URL = r'''(?x:
4592 https?://
4593 (?:\w+\.)?
4594 (?:
4595 youtube(?:kids)?\.com|
4596 %(invidious)s
4597 )/
4598 (?:
4599 (?P<channel_type>channel|c|user|browse)/|
4600 (?P<not_channel>
4601 feed/|hashtag/|
4602 (?:playlist|watch)\?.*?\blist=
4603 )|
4604 (?!(?:%(reserved_names)s)\b) # Direct URLs
4605 )
4606 (?P<id>[^/?\#&]+)
4607 )''' % {
4608 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
4609 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4610 }
4611 IE_NAME = 'youtube:tab'
4612
4613 _TESTS = [{
4614 'note': 'playlists, multipage',
4615 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
4616 'playlist_mincount': 94,
4617 'info_dict': {
4618 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4619 'title': 'Igor Kleiner - Playlists',
4620 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4621 'uploader': 'Igor Kleiner',
4622 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4623 'channel': 'Igor Kleiner',
4624 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4625 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4626 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4627 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4628 'channel_follower_count': int
4629 },
4630 }, {
4631 'note': 'playlists, multipage, different order',
4632 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
4633 'playlist_mincount': 94,
4634 'info_dict': {
4635 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
4636 'title': 'Igor Kleiner - Playlists',
4637 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
4638 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4639 'uploader': 'Igor Kleiner',
4640 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4641 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
4642 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
4643 'channel': 'Igor Kleiner',
4644 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
4645 'channel_follower_count': int
4646 },
4647 }, {
4648 'note': 'playlists, series',
4649 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
4650 'playlist_mincount': 5,
4651 'info_dict': {
4652 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4653 'title': '3Blue1Brown - Playlists',
4654 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4655 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4656 'uploader': '3Blue1Brown',
4657 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4658 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4659 'channel': '3Blue1Brown',
4660 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4661 'tags': ['Mathematics'],
4662 'channel_follower_count': int
4663 },
4664 }, {
4665 'note': 'playlists, singlepage',
4666 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
4667 'playlist_mincount': 4,
4668 'info_dict': {
4669 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4670 'title': 'ThirstForScience - Playlists',
4671 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
4672 'uploader': 'ThirstForScience',
4673 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4674 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4675 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
4676 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
4677 'tags': 'count:13',
4678 'channel': 'ThirstForScience',
4679 'channel_follower_count': int
4680 }
4681 }, {
4682 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
4683 'only_matching': True,
4684 }, {
4685 'note': 'basic, single video playlist',
4686 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4687 'info_dict': {
4688 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4689 'uploader': 'Sergey M.',
4690 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4691 'title': 'youtube-dl public playlist',
4692 'description': '',
4693 'tags': [],
4694 'view_count': int,
4695 'modified_date': '20201130',
4696 'channel': 'Sergey M.',
4697 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4698 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4699 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4700 },
4701 'playlist_count': 1,
4702 }, {
4703 'note': 'empty playlist',
4704 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4705 'info_dict': {
4706 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4707 'uploader': 'Sergey M.',
4708 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
4709 'title': 'youtube-dl empty playlist',
4710 'tags': [],
4711 'channel': 'Sergey M.',
4712 'description': '',
4713 'modified_date': '20160902',
4714 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
4715 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4716 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4717 },
4718 'playlist_count': 0,
4719 }, {
4720 'note': 'Home tab',
4721 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
4722 'info_dict': {
4723 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4724 'title': 'lex will - Home',
4725 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4726 'uploader': 'lex will',
4727 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4728 'channel': 'lex will',
4729 'tags': ['bible', 'history', 'prophesy'],
4730 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4731 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4732 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4733 'channel_follower_count': int
4734 },
4735 'playlist_mincount': 2,
4736 }, {
4737 'note': 'Videos tab',
4738 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
4739 'info_dict': {
4740 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4741 'title': 'lex will - Videos',
4742 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4743 'uploader': 'lex will',
4744 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4745 'tags': ['bible', 'history', 'prophesy'],
4746 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4747 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4748 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4749 'channel': 'lex will',
4750 'channel_follower_count': int
4751 },
4752 'playlist_mincount': 975,
4753 }, {
4754 'note': 'Videos tab, sorted by popular',
4755 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
4756 'info_dict': {
4757 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4758 'title': 'lex will - Videos',
4759 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4760 'uploader': 'lex will',
4761 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4762 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4763 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4764 'channel': 'lex will',
4765 'tags': ['bible', 'history', 'prophesy'],
4766 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4767 'channel_follower_count': int
4768 },
4769 'playlist_mincount': 199,
4770 }, {
4771 'note': 'Playlists tab',
4772 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
4773 'info_dict': {
4774 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4775 'title': 'lex will - Playlists',
4776 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4777 'uploader': 'lex will',
4778 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4779 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4780 'channel': 'lex will',
4781 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4782 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4783 'tags': ['bible', 'history', 'prophesy'],
4784 'channel_follower_count': int
4785 },
4786 'playlist_mincount': 17,
4787 }, {
4788 'note': 'Community tab',
4789 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
4790 'info_dict': {
4791 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4792 'title': 'lex will - Community',
4793 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4794 'uploader': 'lex will',
4795 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4796 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4797 'channel': 'lex will',
4798 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4799 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4800 'tags': ['bible', 'history', 'prophesy'],
4801 'channel_follower_count': int
4802 },
4803 'playlist_mincount': 18,
4804 }, {
4805 'note': 'Channels tab',
4806 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
4807 'info_dict': {
4808 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4809 'title': 'lex will - Channels',
4810 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
4811 'uploader': 'lex will',
4812 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4813 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4814 'channel': 'lex will',
4815 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
4816 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
4817 'tags': ['bible', 'history', 'prophesy'],
4818 'channel_follower_count': int
4819 },
4820 'playlist_mincount': 12,
4821 }, {
4822 'note': 'Search tab',
4823 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
4824 'playlist_mincount': 40,
4825 'info_dict': {
4826 'id': 'UCYO_jab_esuFRV4b17AJtAw',
4827 'title': '3Blue1Brown - Search - linear algebra',
4828 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
4829 'uploader': '3Blue1Brown',
4830 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
4831 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4832 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
4833 'tags': ['Mathematics'],
4834 'channel': '3Blue1Brown',
4835 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
4836 'channel_follower_count': int
4837 },
4838 }, {
4839 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4840 'only_matching': True,
4841 }, {
4842 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4843 'only_matching': True,
4844 }, {
4845 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
4846 'only_matching': True,
4847 }, {
4848 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
4849 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4850 'info_dict': {
4851 'title': '29C3: Not my department',
4852 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
4853 'uploader': 'Christiaan008',
4854 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4855 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
4856 'tags': [],
4857 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4858 'view_count': int,
4859 'modified_date': '20150605',
4860 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
4861 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
4862 'channel': 'Christiaan008',
4863 },
4864 'playlist_count': 96,
4865 }, {
4866 'note': 'Large playlist',
4867 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
4868 'info_dict': {
4869 'title': 'Uploads from Cauchemar',
4870 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
4871 'uploader': 'Cauchemar',
4872 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4873 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
4874 'tags': [],
4875 'modified_date': r're:\d{8}',
4876 'channel': 'Cauchemar',
4877 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
4878 'view_count': int,
4879 'description': '',
4880 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
4881 },
4882 'playlist_mincount': 1123,
4883 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4884 }, {
4885 'note': 'even larger playlist, 8832 videos',
4886 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
4887 'only_matching': True,
4888 }, {
4889 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
4890 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
4891 'info_dict': {
4892 'title': 'Uploads from Interstellar Movie',
4893 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4894 'uploader': 'Interstellar Movie',
4895 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4896 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
4897 'tags': [],
4898 'view_count': int,
4899 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
4900 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
4901 'channel': 'Interstellar Movie',
4902 'description': '',
4903 'modified_date': r're:\d{8}',
4904 },
4905 'playlist_mincount': 21,
4906 }, {
4907 'note': 'Playlist with "show unavailable videos" button',
4908 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
4909 'info_dict': {
4910 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
4911 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
4912 'uploader': 'Phim Siêu Nhân Nhật Bản',
4913 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4914 'view_count': int,
4915 'channel': 'Phim Siêu Nhân Nhật Bản',
4916 'tags': [],
4917 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4918 'description': '',
4919 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
4920 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
4921 'modified_date': r're:\d{8}',
4922 },
4923 'playlist_mincount': 200,
4924 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4925 }, {
4926 'note': 'Playlist with unavailable videos in page 7',
4927 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
4928 'info_dict': {
4929 'title': 'Uploads from BlankTV',
4930 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
4931 'uploader': 'BlankTV',
4932 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4933 'channel': 'BlankTV',
4934 'channel_url': 'https://www.youtube.com/c/blanktv',
4935 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
4936 'view_count': int,
4937 'tags': [],
4938 'uploader_url': 'https://www.youtube.com/c/blanktv',
4939 'modified_date': r're:\d{8}',
4940 'description': '',
4941 },
4942 'playlist_mincount': 1000,
4943 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
4944 }, {
4945 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
4946 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4947 'info_dict': {
4948 'title': 'Data Analysis with Dr Mike Pound',
4949 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
4950 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4951 'uploader': 'Computerphile',
4952 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
4953 'uploader_url': 'https://www.youtube.com/user/Computerphile',
4954 'tags': [],
4955 'view_count': int,
4956 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
4957 'channel_url': 'https://www.youtube.com/user/Computerphile',
4958 'channel': 'Computerphile',
4959 },
4960 'playlist_mincount': 11,
4961 }, {
4962 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
4963 'only_matching': True,
4964 }, {
4965 'note': 'Playlist URL that does not actually serve a playlist',
4966 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
4967 'info_dict': {
4968 'id': 'FqZTN594JQw',
4969 'ext': 'webm',
4970 'title': "Smiley's People 01 detective, Adventure Series, Action",
4971 'uploader': 'STREEM',
4972 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
4973 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
4974 'upload_date': '20150526',
4975 'license': 'Standard YouTube License',
4976 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
4977 'categories': ['People & Blogs'],
4978 'tags': list,
4979 'view_count': int,
4980 'like_count': int,
4981 },
4982 'params': {
4983 'skip_download': True,
4984 },
4985 'skip': 'This video is not available.',
4986 'add_ie': [YoutubeIE.ie_key()],
4987 }, {
4988 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
4989 'only_matching': True,
4990 }, {
4991 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
4992 'only_matching': True,
4993 }, {
4994 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
4995 'info_dict': {
4996 'id': 'GgL890LIznQ', # This will keep changing
4997 'ext': 'mp4',
4998 'title': str,
4999 'uploader': 'Sky News',
5000 'uploader_id': 'skynews',
5001 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
5002 'upload_date': r're:\d{8}',
5003 'description': str,
5004 'categories': ['News & Politics'],
5005 'tags': list,
5006 'like_count': int,
5007 'release_timestamp': 1642502819,
5008 'channel': 'Sky News',
5009 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5010 'age_limit': 0,
5011 'view_count': int,
5012 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
5013 'playable_in_embed': True,
5014 'release_date': '20220118',
5015 'availability': 'public',
5016 'live_status': 'is_live',
5017 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5018 'channel_follower_count': int
5019 },
5020 'params': {
5021 'skip_download': True,
5022 },
5023 'expected_warnings': ['Ignoring subtitle tracks found in '],
5024 }, {
5025 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5026 'info_dict': {
5027 'id': 'a48o2S1cPoo',
5028 'ext': 'mp4',
5029 'title': 'The Young Turks - Live Main Show',
5030 'uploader': 'The Young Turks',
5031 'uploader_id': 'TheYoungTurks',
5032 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
5033 'upload_date': '20150715',
5034 'license': 'Standard YouTube License',
5035 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5036 'categories': ['News & Politics'],
5037 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5038 'like_count': int,
5039 },
5040 'params': {
5041 'skip_download': True,
5042 },
5043 'only_matching': True,
5044 }, {
5045 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5046 'only_matching': True,
5047 }, {
5048 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5049 'only_matching': True,
5050 }, {
5051 'note': 'A channel that is not live. Should raise error',
5052 'url': 'https://www.youtube.com/user/numberphile/live',
5053 'only_matching': True,
5054 }, {
5055 'url': 'https://www.youtube.com/feed/trending',
5056 'only_matching': True,
5057 }, {
5058 'url': 'https://www.youtube.com/feed/library',
5059 'only_matching': True,
5060 }, {
5061 'url': 'https://www.youtube.com/feed/history',
5062 'only_matching': True,
5063 }, {
5064 'url': 'https://www.youtube.com/feed/subscriptions',
5065 'only_matching': True,
5066 }, {
5067 'url': 'https://www.youtube.com/feed/watch_later',
5068 'only_matching': True,
5069 }, {
5070 'note': 'Recommended - redirects to home page.',
5071 'url': 'https://www.youtube.com/feed/recommended',
5072 'only_matching': True,
5073 }, {
5074 'note': 'inline playlist with not always working continuations',
5075 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5076 'only_matching': True,
5077 }, {
5078 'url': 'https://www.youtube.com/course',
5079 'only_matching': True,
5080 }, {
5081 'url': 'https://www.youtube.com/zsecurity',
5082 'only_matching': True,
5083 }, {
5084 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5085 'only_matching': True,
5086 }, {
5087 'url': 'https://www.youtube.com/TheYoungTurks/live',
5088 'only_matching': True,
5089 }, {
5090 'url': 'https://www.youtube.com/hashtag/cctv9',
5091 'info_dict': {
5092 'id': 'cctv9',
5093 'title': '#cctv9',
5094 'tags': [],
5095 },
5096 'playlist_mincount': 350,
5097 }, {
5098 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5099 'only_matching': True,
5100 }, {
5101 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5102 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5103 'only_matching': True
5104 }, {
5105 'note': '/browse/ should redirect to /channel/',
5106 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5107 'only_matching': True
5108 }, {
5109 'note': 'VLPL, should redirect to playlist?list=PL...',
5110 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5111 'info_dict': {
5112 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5113 'uploader': 'NoCopyrightSounds',
5114 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5115 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5116 'title': 'NCS Releases',
5117 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5118 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5119 'modified_date': r're:\d{8}',
5120 'view_count': int,
5121 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5122 'tags': [],
5123 'channel': 'NoCopyrightSounds',
5124 },
5125 'playlist_mincount': 166,
5126 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5127 }, {
5128 'note': 'Topic, should redirect to playlist?list=UU...',
5129 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5130 'info_dict': {
5131 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5132 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5133 'title': 'Uploads from Royalty Free Music - Topic',
5134 'uploader': 'Royalty Free Music - Topic',
5135 'tags': [],
5136 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5137 'channel': 'Royalty Free Music - Topic',
5138 'view_count': int,
5139 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5140 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5141 'modified_date': r're:\d{8}',
5142 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5143 'description': '',
5144 },
5145 'expected_warnings': [
5146 'The URL does not have a videos tab',
5147 r'[Uu]navailable videos (are|will be) hidden',
5148 ],
5149 'playlist_mincount': 101,
5150 }, {
5151 'note': 'Topic without a UU playlist',
5152 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5153 'info_dict': {
5154 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5155 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5156 'tags': [],
5157 },
5158 'expected_warnings': [
5159 'the playlist redirect gave error',
5160 ],
5161 'playlist_mincount': 9,
5162 }, {
5163 'note': 'Youtube music Album',
5164 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5165 'info_dict': {
5166 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5167 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5168 'tags': [],
5169 'view_count': int,
5170 'description': '',
5171 'availability': 'unlisted',
5172 'modified_date': r're:\d{8}',
5173 },
5174 'playlist_count': 50,
5175 }, {
5176 'note': 'unlisted single video playlist',
5177 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5178 'info_dict': {
5179 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5180 'uploader': 'colethedj',
5181 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5182 'title': 'yt-dlp unlisted playlist test',
5183 'availability': 'unlisted',
5184 'tags': [],
5185 'modified_date': '20211208',
5186 'channel': 'colethedj',
5187 'view_count': int,
5188 'description': '',
5189 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5190 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5191 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5192 },
5193 'playlist_count': 1,
5194 }, {
5195 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5196 'url': 'https://www.youtube.com/feed/recommended',
5197 'info_dict': {
5198 'id': 'recommended',
5199 'title': 'recommended',
5200 'tags': [],
5201 },
5202 'playlist_mincount': 50,
5203 'params': {
5204 'skip_download': True,
5205 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5206 },
5207 }, {
5208 'note': 'API Fallback: /videos tab, sorted by oldest first',
5209 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5210 'info_dict': {
5211 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5212 'title': 'Cody\'sLab - Videos',
5213 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5214 'uploader': 'Cody\'sLab',
5215 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5216 'channel': 'Cody\'sLab',
5217 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5218 'tags': [],
5219 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5220 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5221 'channel_follower_count': int
5222 },
5223 'playlist_mincount': 650,
5224 'params': {
5225 'skip_download': True,
5226 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5227 },
5228 }, {
5229 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5230 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5231 'info_dict': {
5232 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5233 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5234 'title': 'Uploads from Royalty Free Music - Topic',
5235 'uploader': 'Royalty Free Music - Topic',
5236 'modified_date': r're:\d{8}',
5237 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5238 'description': '',
5239 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5240 'tags': [],
5241 'channel': 'Royalty Free Music - Topic',
5242 'view_count': int,
5243 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5244 },
5245 'expected_warnings': [
5246 'does not have a videos tab',
5247 r'[Uu]navailable videos (are|will be) hidden',
5248 ],
5249 'playlist_mincount': 101,
5250 'params': {
5251 'skip_download': True,
5252 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5253 },
5254 }, {
5255 'note': 'non-standard redirect to regional channel',
5256 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5257 'only_matching': True
5258 }, {
5259 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5260 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5261 'info_dict': {
5262 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5263 'modified_date': '20220407',
5264 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5265 'tags': [],
5266 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5267 'uploader': 'pukkandan',
5268 'availability': 'unlisted',
5269 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5270 'channel': 'pukkandan',
5271 'description': 'Test for collaborative playlist',
5272 'title': 'yt-dlp test - collaborative playlist',
5273 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5274 },
5275 'playlist_mincount': 2
5276 }]
5277
5278 @classmethod
5279 def suitable(cls, url):
5280 return False if YoutubeIE.suitable(url) else super().suitable(url)
5281
5282 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
5283
5284 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
5285 def _real_extract(self, url, smuggled_data):
5286 item_id = self._match_id(url)
5287 url = compat_urlparse.urlunparse(
5288 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
5289 compat_opts = self.get_param('compat_opts', [])
5290
5291 def get_mobj(url):
5292 mobj = self._URL_RE.match(url).groupdict()
5293 mobj.update((k, '') for k, v in mobj.items() if v is None)
5294 return mobj
5295
5296 mobj, redirect_warning = get_mobj(url), None
5297 # Youtube returns incomplete data if tabname is not lower case
5298 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
5299 if is_channel:
5300 if smuggled_data.get('is_music_url'):
5301 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
5302 item_id = item_id[2:]
5303 pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={item_id}', '', '', False
5304 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
5305 mdata = self._extract_tab_endpoint(
5306 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
5307 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
5308 get_all=False, expected_type=compat_str)
5309 if not murl:
5310 raise ExtractorError('Failed to resolve album to playlist')
5311 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
5312 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
5313 pre = f'https://www.youtube.com/channel/{item_id}'
5314
5315 original_tab_name = tab
5316 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
5317 # Home URLs should redirect to /videos/
5318 redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
5319 'To download only the videos in the home page, add a "/featured" to the URL')
5320 tab = '/videos'
5321
5322 url = ''.join((pre, tab, post))
5323 mobj = get_mobj(url)
5324
5325 # Handle both video/playlist URLs
5326 qs = parse_qs(url)
5327 video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
5328
5329 if not video_id and mobj['not_channel'].startswith('watch'):
5330 if not playlist_id:
5331 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
5332 raise ExtractorError('Unable to recognize tab page')
5333 # Common mistake: https://www.youtube.com/watch?list=playlist_id
5334 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
5335 url = f'https://www.youtube.com/playlist?list={playlist_id}'
5336 mobj = get_mobj(url)
5337
5338 if video_id and playlist_id:
5339 if self.get_param('noplaylist'):
5340 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
5341 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5342 ie=YoutubeIE.ie_key(), video_id=video_id)
5343 self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}')
5344
5345 data, ytcfg = self._extract_data(url, item_id)
5346
5347 # YouTube may provide a non-standard redirect to the regional channel
5348 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
5349 redirect_url = traverse_obj(
5350 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
5351 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
5352 redirect_url = ''.join((
5353 urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
5354 self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
5355 return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
5356
5357 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5358 if tabs:
5359 selected_tab = self._extract_selected_tab(tabs)
5360 selected_tab_name = selected_tab.get('title', '').lower()
5361 if selected_tab_name == 'home':
5362 selected_tab_name = 'featured'
5363 requested_tab_name = mobj['tab'][1:]
5364 if 'no-youtube-channel-redirect' not in compat_opts:
5365 if requested_tab_name == 'live':
5366 # Live tab should have redirected to the video
5367 raise ExtractorError('The channel is not currently live', expected=True)
5368 if requested_tab_name not in ('', selected_tab_name):
5369 redirect_warning = f'The channel does not have a {requested_tab_name} tab'
5370 if not original_tab_name:
5371 if item_id[:2] == 'UC':
5372 # Topic channels don't have /videos. Use the equivalent playlist instead
5373 pl_id = f'UU{item_id[2:]}'
5374 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
5375 try:
5376 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
5377 except ExtractorError:
5378 redirect_warning += ' and the playlist redirect gave error'
5379 else:
5380 item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
5381 redirect_warning += f'. Redirecting to playlist {pl_id} instead'
5382 if selected_tab_name and selected_tab_name != requested_tab_name:
5383 redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
5384 else:
5385 raise ExtractorError(redirect_warning, expected=True)
5386
5387 if redirect_warning:
5388 self.to_screen(redirect_warning)
5389 self.write_debug(f'Final URL: {url}')
5390
5391 # YouTube sometimes provides a button to reload playlist with unavailable videos.
5392 if 'no-youtube-unavailable-videos' not in compat_opts:
5393 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
5394 self._extract_and_report_alerts(data, only_once=True)
5395 tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
5396 if tabs:
5397 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
5398
5399 playlist = traverse_obj(
5400 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
5401 if playlist:
5402 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
5403
5404 video_id = traverse_obj(
5405 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
5406 if video_id:
5407 if mobj['tab'] != '/live': # live tab is expected to redirect to video
5408 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
5409 return self.url_result(f'https://www.youtube.com/watch?v={video_id}',
5410 ie=YoutubeIE.ie_key(), video_id=video_id)
5411
5412 raise ExtractorError('Unable to recognize tab page')
5413
5414
5415class YoutubePlaylistIE(InfoExtractor):
5416 IE_DESC = 'YouTube playlists'
5417 _VALID_URL = r'''(?x)(?:
5418 (?:https?://)?
5419 (?:\w+\.)?
5420 (?:
5421 (?:
5422 youtube(?:kids)?\.com|
5423 %(invidious)s
5424 )
5425 /.*?\?.*?\blist=
5426 )?
5427 (?P<id>%(playlist_id)s)
5428 )''' % {
5429 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
5430 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5431 }
5432 IE_NAME = 'youtube:playlist'
5433 _TESTS = [{
5434 'note': 'issue #673',
5435 'url': 'PLBB231211A4F62143',
5436 'info_dict': {
5437 'title': '[OLD]Team Fortress 2 (Class-based LP)',
5438 'id': 'PLBB231211A4F62143',
5439 'uploader': 'Wickman',
5440 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5441 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
5442 'view_count': int,
5443 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
5444 'modified_date': r're:\d{8}',
5445 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
5446 'channel': 'Wickman',
5447 'tags': [],
5448 'channel_url': 'https://www.youtube.com/user/Wickydoo',
5449 },
5450 'playlist_mincount': 29,
5451 }, {
5452 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5453 'info_dict': {
5454 'title': 'YDL_safe_search',
5455 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
5456 },
5457 'playlist_count': 2,
5458 'skip': 'This playlist is private',
5459 }, {
5460 'note': 'embedded',
5461 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5462 'playlist_count': 4,
5463 'info_dict': {
5464 'title': 'JODA15',
5465 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
5466 'uploader': 'milan',
5467 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5468 'description': '',
5469 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5470 'tags': [],
5471 'modified_date': '20140919',
5472 'view_count': int,
5473 'channel': 'milan',
5474 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
5475 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
5476 },
5477 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5478 }, {
5479 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5480 'playlist_mincount': 654,
5481 'info_dict': {
5482 'title': '2018 Chinese New Singles (11/6 updated)',
5483 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
5484 'uploader': 'LBK',
5485 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5486 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
5487 'channel': 'LBK',
5488 'view_count': int,
5489 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
5490 'tags': [],
5491 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
5492 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
5493 'modified_date': r're:\d{8}',
5494 },
5495 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5496 }, {
5497 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
5498 'only_matching': True,
5499 }, {
5500 # music album playlist
5501 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
5502 'only_matching': True,
5503 }]
5504
5505 @classmethod
5506 def suitable(cls, url):
5507 if YoutubeTabIE.suitable(url):
5508 return False
5509 from ..utils import parse_qs
5510 qs = parse_qs(url)
5511 if qs.get('v', [None])[0]:
5512 return False
5513 return super().suitable(url)
5514
5515 def _real_extract(self, url):
5516 playlist_id = self._match_id(url)
5517 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
5518 url = update_url_query(
5519 'https://www.youtube.com/playlist',
5520 parse_qs(url) or {'list': playlist_id})
5521 if is_music_url:
5522 url = smuggle_url(url, {'is_music_url': True})
5523 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5524
5525
5526class YoutubeYtBeIE(InfoExtractor):
5527 IE_DESC = 'youtu.be'
5528 _VALID_URL = r'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{11})/*?.*?\blist=(?P<playlist_id>%(playlist_id)s)' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
5529 _TESTS = [{
5530 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
5531 'info_dict': {
5532 'id': 'yeWKywCrFtk',
5533 'ext': 'mp4',
5534 'title': 'Small Scale Baler and Braiding Rugs',
5535 'uploader': 'Backus-Page House Museum',
5536 'uploader_id': 'backuspagemuseum',
5537 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
5538 'upload_date': '20161008',
5539 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
5540 'categories': ['Nonprofits & Activism'],
5541 'tags': list,
5542 'like_count': int,
5543 'age_limit': 0,
5544 'playable_in_embed': True,
5545 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
5546 'channel': 'Backus-Page House Museum',
5547 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
5548 'live_status': 'not_live',
5549 'view_count': int,
5550 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
5551 'availability': 'public',
5552 'duration': 59,
5553 },
5554 'params': {
5555 'noplaylist': True,
5556 'skip_download': True,
5557 },
5558 }, {
5559 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
5560 'only_matching': True,
5561 }]
5562
5563 def _real_extract(self, url):
5564 mobj = self._match_valid_url(url)
5565 video_id = mobj.group('id')
5566 playlist_id = mobj.group('playlist_id')
5567 return self.url_result(
5568 update_url_query('https://www.youtube.com/watch', {
5569 'v': video_id,
5570 'list': playlist_id,
5571 'feature': 'youtu.be',
5572 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5573
5574
5575class YoutubeLivestreamEmbedIE(InfoExtractor):
5576 IE_DESC = 'YouTube livestream embeds'
5577 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
5578 _TESTS = [{
5579 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
5580 'only_matching': True,
5581 }]
5582
5583 def _real_extract(self, url):
5584 channel_id = self._match_id(url)
5585 return self.url_result(
5586 f'https://www.youtube.com/channel/{channel_id}/live',
5587 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
5588
5589
5590class YoutubeYtUserIE(InfoExtractor):
5591 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
5592 IE_NAME = 'youtube:user'
5593 _VALID_URL = r'ytuser:(?P<id>.+)'
5594 _TESTS = [{
5595 'url': 'ytuser:phihag',
5596 'only_matching': True,
5597 }]
5598
5599 def _real_extract(self, url):
5600 user_id = self._match_id(url)
5601 return self.url_result(
5602 'https://www.youtube.com/user/%s/videos' % user_id,
5603 ie=YoutubeTabIE.ie_key(), video_id=user_id)
5604
5605
5606class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
5607 IE_NAME = 'youtube:favorites'
5608 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
5609 _VALID_URL = r':ytfav(?:ou?rite)?s?'
5610 _LOGIN_REQUIRED = True
5611 _TESTS = [{
5612 'url': ':ytfav',
5613 'only_matching': True,
5614 }, {
5615 'url': ':ytfavorites',
5616 'only_matching': True,
5617 }]
5618
5619 def _real_extract(self, url):
5620 return self.url_result(
5621 'https://www.youtube.com/playlist?list=LL',
5622 ie=YoutubeTabIE.ie_key())
5623
5624
5625class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
5626 IE_NAME = 'youtube:notif'
5627 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
5628 _VALID_URL = r':ytnotif(?:ication)?s?'
5629 _LOGIN_REQUIRED = True
5630 _TESTS = [{
5631 'url': ':ytnotif',
5632 'only_matching': True,
5633 }, {
5634 'url': ':ytnotifications',
5635 'only_matching': True,
5636 }]
5637
5638 def _extract_notification_menu(self, response, continuation_list):
5639 notification_list = traverse_obj(
5640 response,
5641 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
5642 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
5643 expected_type=list) or []
5644 continuation_list[0] = None
5645 for item in notification_list:
5646 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
5647 if entry:
5648 yield entry
5649 continuation = item.get('continuationItemRenderer')
5650 if continuation:
5651 continuation_list[0] = continuation
5652
5653 def _extract_notification_renderer(self, notification):
5654 video_id = traverse_obj(
5655 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
5656 url = f'https://www.youtube.com/watch?v={video_id}'
5657 channel_id = None
5658 if not video_id:
5659 browse_ep = traverse_obj(
5660 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
5661 channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
5662 post_id = self._search_regex(
5663 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
5664 'post id', default=None)
5665 if not channel_id or not post_id:
5666 return
5667 # The direct /post url redirects to this in the browser
5668 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
5669
5670 channel = traverse_obj(
5671 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
5672 expected_type=str)
5673 title = self._search_regex(
5674 rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
5675 'video title', default=None)
5676 if title:
5677 title = title.replace('\xad', '') # remove soft hyphens
5678 upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
5679 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
5680 else None)
5681 return {
5682 '_type': 'url',
5683 'url': url,
5684 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
5685 'video_id': video_id,
5686 'title': title,
5687 'channel_id': channel_id,
5688 'channel': channel,
5689 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
5690 'upload_date': upload_date,
5691 }
5692
5693 def _notification_menu_entries(self, ytcfg):
5694 continuation_list = [None]
5695 response = None
5696 for page in itertools.count(1):
5697 ctoken = traverse_obj(
5698 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
5699 response = self._extract_response(
5700 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
5701 ep='notification/get_notification_menu', check_get_keys='actions',
5702 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
5703 yield from self._extract_notification_menu(response, continuation_list)
5704 if not continuation_list[0]:
5705 break
5706
5707 def _real_extract(self, url):
5708 display_id = 'notifications'
5709 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
5710 self._report_playlist_authcheck(ytcfg)
5711 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
5712
5713
5714class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5715 IE_DESC = 'YouTube search'
5716 IE_NAME = 'youtube:search'
5717 _SEARCH_KEY = 'ytsearch'
5718 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
5719 _TESTS = [{
5720 'url': 'ytsearch5:youtube-dl test video',
5721 'playlist_count': 5,
5722 'info_dict': {
5723 'id': 'youtube-dl test video',
5724 'title': 'youtube-dl test video',
5725 }
5726 }]
5727
5728
5729class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
5730 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
5731 _SEARCH_KEY = 'ytsearchdate'
5732 IE_DESC = 'YouTube search, newest videos first'
5733 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
5734 _TESTS = [{
5735 'url': 'ytsearchdate5:youtube-dl test video',
5736 'playlist_count': 5,
5737 'info_dict': {
5738 'id': 'youtube-dl test video',
5739 'title': 'youtube-dl test video',
5740 }
5741 }]
5742
5743
5744class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
5745 IE_DESC = 'YouTube search URLs with sorting and filter support'
5746 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
5747 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5748 _TESTS = [{
5749 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
5750 'playlist_mincount': 5,
5751 'info_dict': {
5752 'id': 'youtube-dl test video',
5753 'title': 'youtube-dl test video',
5754 }
5755 }, {
5756 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
5757 'playlist_mincount': 5,
5758 'info_dict': {
5759 'id': 'python',
5760 'title': 'python',
5761 }
5762 }, {
5763 'url': 'https://www.youtube.com/results?search_query=%23cats',
5764 'playlist_mincount': 1,
5765 'info_dict': {
5766 'id': '#cats',
5767 'title': '#cats',
5768 'entries': [{
5769 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
5770 'title': '#cats',
5771 }],
5772 },
5773 }, {
5774 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
5775 'only_matching': True,
5776 }]
5777
5778 def _real_extract(self, url):
5779 qs = parse_qs(url)
5780 query = (qs.get('search_query') or qs.get('q'))[0]
5781 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
5782
5783
5784class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
5785 IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
5786 IE_NAME = 'youtube:music:search_url'
5787 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
5788 _TESTS = [{
5789 'url': 'https://music.youtube.com/search?q=royalty+free+music',
5790 'playlist_count': 16,
5791 'info_dict': {
5792 'id': 'royalty free music',
5793 'title': 'royalty free music',
5794 }
5795 }, {
5796 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
5797 'playlist_mincount': 30,
5798 'info_dict': {
5799 'id': 'royalty free music - songs',
5800 'title': 'royalty free music - songs',
5801 },
5802 'params': {'extract_flat': 'in_playlist'}
5803 }, {
5804 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
5805 'playlist_mincount': 30,
5806 'info_dict': {
5807 'id': 'royalty free music - community playlists',
5808 'title': 'royalty free music - community playlists',
5809 },
5810 'params': {'extract_flat': 'in_playlist'}
5811 }]
5812
5813 _SECTIONS = {
5814 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
5815 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
5816 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
5817 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
5818 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
5819 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
5820 }
5821
5822 def _real_extract(self, url):
5823 qs = parse_qs(url)
5824 query = (qs.get('search_query') or qs.get('q'))[0]
5825 params = qs.get('sp', (None,))[0]
5826 if params:
5827 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
5828 else:
5829 section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
5830 params = self._SECTIONS.get(section)
5831 if not params:
5832 section = None
5833 title = join_nonempty(query, section, delim=' - ')
5834 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
5835
5836
5837class YoutubeFeedsInfoExtractor(InfoExtractor):
5838 """
5839 Base class for feed extractors
5840 Subclasses must re-define the _FEED_NAME property.
5841 """
5842 _LOGIN_REQUIRED = True
5843 _FEED_NAME = 'feeds'
5844
5845 def _real_initialize(self):
5846 YoutubeBaseInfoExtractor._check_login_required(self)
5847
5848 @classproperty
5849 def IE_NAME(self):
5850 return f'youtube:{self._FEED_NAME}'
5851
5852 def _real_extract(self, url):
5853 return self.url_result(
5854 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
5855
5856
5857class YoutubeWatchLaterIE(InfoExtractor):
5858 IE_NAME = 'youtube:watchlater'
5859 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
5860 _VALID_URL = r':ytwatchlater'
5861 _TESTS = [{
5862 'url': ':ytwatchlater',
5863 'only_matching': True,
5864 }]
5865
5866 def _real_extract(self, url):
5867 return self.url_result(
5868 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
5869
5870
5871class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
5872 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
5873 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
5874 _FEED_NAME = 'recommended'
5875 _LOGIN_REQUIRED = False
5876 _TESTS = [{
5877 'url': ':ytrec',
5878 'only_matching': True,
5879 }, {
5880 'url': ':ytrecommended',
5881 'only_matching': True,
5882 }, {
5883 'url': 'https://youtube.com',
5884 'only_matching': True,
5885 }]
5886
5887
5888class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
5889 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
5890 _VALID_URL = r':ytsub(?:scription)?s?'
5891 _FEED_NAME = 'subscriptions'
5892 _TESTS = [{
5893 'url': ':ytsubs',
5894 'only_matching': True,
5895 }, {
5896 'url': ':ytsubscriptions',
5897 'only_matching': True,
5898 }]
5899
5900
5901class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
5902 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
5903 _VALID_URL = r':ythis(?:tory)?'
5904 _FEED_NAME = 'history'
5905 _TESTS = [{
5906 'url': ':ythistory',
5907 'only_matching': True,
5908 }]
5909
5910
5911class YoutubeStoriesIE(InfoExtractor):
5912 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
5913 IE_NAME = 'youtube:stories'
5914 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
5915 _TESTS = [{
5916 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
5917 'only_matching': True,
5918 }]
5919
5920 def _real_extract(self, url):
5921 playlist_id = f'RLTD{self._match_id(url)}'
5922 return self.url_result(
5923 f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1',
5924 ie=YoutubeTabIE, video_id=playlist_id)
5925
5926
5927class YoutubeTruncatedURLIE(InfoExtractor):
5928 IE_NAME = 'youtube:truncated_url'
5929 IE_DESC = False # Do not list
5930 _VALID_URL = r'''(?x)
5931 (?:https?://)?
5932 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
5933 (?:watch\?(?:
5934 feature=[a-z_]+|
5935 annotation_id=annotation_[^&]+|
5936 x-yt-cl=[0-9]+|
5937 hl=[^&]*|
5938 t=[0-9]+
5939 )?
5940 |
5941 attribution_link\?a=[^&]+
5942 )
5943 $
5944 '''
5945
5946 _TESTS = [{
5947 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
5948 'only_matching': True,
5949 }, {
5950 'url': 'https://www.youtube.com/watch?',
5951 'only_matching': True,
5952 }, {
5953 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
5954 'only_matching': True,
5955 }, {
5956 'url': 'https://www.youtube.com/watch?feature=foo',
5957 'only_matching': True,
5958 }, {
5959 'url': 'https://www.youtube.com/watch?hl=en-GB',
5960 'only_matching': True,
5961 }, {
5962 'url': 'https://www.youtube.com/watch?t=2372',
5963 'only_matching': True,
5964 }]
5965
5966 def _real_extract(self, url):
5967 raise ExtractorError(
5968 'Did you forget to quote the URL? Remember that & is a meta '
5969 'character in most shells, so you want to put the URL in quotes, '
5970 'like youtube-dl '
5971 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
5972 ' or simply youtube-dl BaW_jenozKc .',
5973 expected=True)
5974
5975
5976class YoutubeClipIE(InfoExtractor):
5977 IE_NAME = 'youtube:clip'
5978 IE_DESC = False # Do not list
5979 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
5980
5981 def _real_extract(self, url):
5982 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
5983 return self.url_result(url, 'Generic')
5984
5985
5986class YoutubeTruncatedIDIE(InfoExtractor):
5987 IE_NAME = 'youtube:truncated_id'
5988 IE_DESC = False # Do not list
5989 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
5990
5991 _TESTS = [{
5992 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
5993 'only_matching': True,
5994 }]
5995
5996 def _real_extract(self, url):
5997 video_id = self._match_id(url)
5998 raise ExtractorError(
5999 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
6000 expected=True)