]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[youtube] Add Invidious list for playlists/channels (#1567)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import base64
6 import calendar
7 import copy
8 import datetime
9 import hashlib
10 import itertools
11 import json
12 import os.path
13 import random
14 import re
15 import time
16 import traceback
17
18 from .common import InfoExtractor, SearchInfoExtractor
19 from ..compat import (
20 compat_chr,
21 compat_HTTPError,
22 compat_parse_qs,
23 compat_str,
24 compat_urllib_parse_unquote_plus,
25 compat_urllib_parse_urlencode,
26 compat_urllib_parse_urlparse,
27 compat_urlparse,
28 )
29 from ..jsinterp import JSInterpreter
30 from ..utils import (
31 bytes_to_intlist,
32 clean_html,
33 datetime_from_str,
34 dict_get,
35 error_to_compat_str,
36 ExtractorError,
37 float_or_none,
38 format_field,
39 int_or_none,
40 intlist_to_bytes,
41 is_html,
42 join_nonempty,
43 mimetype2ext,
44 network_exceptions,
45 orderedSet,
46 parse_codecs,
47 parse_count,
48 parse_duration,
49 parse_iso8601,
50 parse_qs,
51 qualities,
52 remove_end,
53 remove_start,
54 smuggle_url,
55 str_or_none,
56 str_to_int,
57 traverse_obj,
58 try_get,
59 unescapeHTML,
60 unified_strdate,
61 unsmuggle_url,
62 update_url_query,
63 url_or_none,
64 urljoin,
65 variadic,
66 )
67
68
69 # any clients starting with _ cannot be explicity requested by the user
70 INNERTUBE_CLIENTS = {
71 'web': {
72 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
73 'INNERTUBE_CONTEXT': {
74 'client': {
75 'clientName': 'WEB',
76 'clientVersion': '2.20210622.10.00',
77 }
78 },
79 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
80 },
81 'web_embedded': {
82 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
83 'INNERTUBE_CONTEXT': {
84 'client': {
85 'clientName': 'WEB_EMBEDDED_PLAYER',
86 'clientVersion': '1.20210620.0.1',
87 },
88 },
89 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
90 },
91 'web_music': {
92 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
93 'INNERTUBE_HOST': 'music.youtube.com',
94 'INNERTUBE_CONTEXT': {
95 'client': {
96 'clientName': 'WEB_REMIX',
97 'clientVersion': '1.20210621.00.00',
98 }
99 },
100 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
101 },
102 'web_creator': {
103 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
104 'INNERTUBE_CONTEXT': {
105 'client': {
106 'clientName': 'WEB_CREATOR',
107 'clientVersion': '1.20210621.00.00',
108 }
109 },
110 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
111 },
112 'android': {
113 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
114 'INNERTUBE_CONTEXT': {
115 'client': {
116 'clientName': 'ANDROID',
117 'clientVersion': '16.20',
118 }
119 },
120 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
121 'REQUIRE_JS_PLAYER': False
122 },
123 'android_embedded': {
124 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
125 'INNERTUBE_CONTEXT': {
126 'client': {
127 'clientName': 'ANDROID_EMBEDDED_PLAYER',
128 'clientVersion': '16.20',
129 },
130 },
131 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
132 'REQUIRE_JS_PLAYER': False
133 },
134 'android_music': {
135 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
136 'INNERTUBE_HOST': 'music.youtube.com',
137 'INNERTUBE_CONTEXT': {
138 'client': {
139 'clientName': 'ANDROID_MUSIC',
140 'clientVersion': '4.32',
141 }
142 },
143 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
144 'REQUIRE_JS_PLAYER': False
145 },
146 'android_creator': {
147 'INNERTUBE_CONTEXT': {
148 'client': {
149 'clientName': 'ANDROID_CREATOR',
150 'clientVersion': '21.24.100',
151 },
152 },
153 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
154 'REQUIRE_JS_PLAYER': False
155 },
156 # ios has HLS live streams
157 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680
158 'ios': {
159 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
160 'INNERTUBE_CONTEXT': {
161 'client': {
162 'clientName': 'IOS',
163 'clientVersion': '16.20',
164 }
165 },
166 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
167 'REQUIRE_JS_PLAYER': False
168 },
169 'ios_embedded': {
170 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8',
171 'INNERTUBE_CONTEXT': {
172 'client': {
173 'clientName': 'IOS_MESSAGES_EXTENSION',
174 'clientVersion': '16.20',
175 },
176 },
177 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
178 'REQUIRE_JS_PLAYER': False
179 },
180 'ios_music': {
181 'INNERTUBE_API_KEY': 'AIzaSyDK3iBpDP9nHVTk2qL73FLJICfOC3c51Og',
182 'INNERTUBE_HOST': 'music.youtube.com',
183 'INNERTUBE_CONTEXT': {
184 'client': {
185 'clientName': 'IOS_MUSIC',
186 'clientVersion': '4.32',
187 },
188 },
189 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
190 'REQUIRE_JS_PLAYER': False
191 },
192 'ios_creator': {
193 'INNERTUBE_CONTEXT': {
194 'client': {
195 'clientName': 'IOS_CREATOR',
196 'clientVersion': '21.24.100',
197 },
198 },
199 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
200 'REQUIRE_JS_PLAYER': False
201 },
202 # mweb has 'ultralow' formats
203 # See: https://github.com/yt-dlp/yt-dlp/pull/557
204 'mweb': {
205 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8',
206 'INNERTUBE_CONTEXT': {
207 'client': {
208 'clientName': 'MWEB',
209 'clientVersion': '2.20210721.07.00',
210 }
211 },
212 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
213 },
214 }
215
216
217 def build_innertube_clients():
218 third_party = {
219 'embedUrl': 'https://google.com', # Can be any valid URL
220 }
221 base_clients = ('android', 'web', 'ios', 'mweb')
222 priority = qualities(base_clients[::-1])
223
224 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
225 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
226 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
227 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
228 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
229 ytcfg['priority'] = 10 * priority(client.split('_', 1)[0])
230
231 if client in base_clients:
232 INNERTUBE_CLIENTS[f'{client}_agegate'] = agegate_ytcfg = copy.deepcopy(ytcfg)
233 agegate_ytcfg['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
234 agegate_ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party
235 agegate_ytcfg['priority'] -= 1
236 elif client.endswith('_embedded'):
237 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party
238 ytcfg['priority'] -= 2
239 else:
240 ytcfg['priority'] -= 3
241
242
243 build_innertube_clients()
244
245
246 class YoutubeBaseInfoExtractor(InfoExtractor):
247 """Provide base functions for Youtube extractors"""
248
249 _RESERVED_NAMES = (
250 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
251 r'shorts|movies|results|shared|hashtag|trending|feed|feeds|'
252 r'browse|oembed|get_video_info|iframe_api|s/player|'
253 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
254
255 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
256
257 _NETRC_MACHINE = 'youtube'
258
259 # If True it will raise an error if no login info is provided
260 _LOGIN_REQUIRED = False
261
262 _INVIDIOUS_SITES = (
263 # invidious-redirect websites
264 r'(?:www\.)?redirect\.invidious\.io',
265 r'(?:(?:www|dev)\.)?invidio\.us',
266 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
267 r'(?:www\.)?invidious\.pussthecat\.org',
268 r'(?:www\.)?invidious\.zee\.li',
269 r'(?:www\.)?invidious\.ethibox\.fr',
270 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
271 # youtube-dl invidious instances list
272 r'(?:(?:www|no)\.)?invidiou\.sh',
273 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
274 r'(?:www\.)?invidious\.kabi\.tk',
275 r'(?:www\.)?invidious\.mastodon\.host',
276 r'(?:www\.)?invidious\.zapashcanon\.fr',
277 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
278 r'(?:www\.)?invidious\.tinfoil-hat\.net',
279 r'(?:www\.)?invidious\.himiko\.cloud',
280 r'(?:www\.)?invidious\.reallyancient\.tech',
281 r'(?:www\.)?invidious\.tube',
282 r'(?:www\.)?invidiou\.site',
283 r'(?:www\.)?invidious\.site',
284 r'(?:www\.)?invidious\.xyz',
285 r'(?:www\.)?invidious\.nixnet\.xyz',
286 r'(?:www\.)?invidious\.048596\.xyz',
287 r'(?:www\.)?invidious\.drycat\.fr',
288 r'(?:www\.)?inv\.skyn3t\.in',
289 r'(?:www\.)?tube\.poal\.co',
290 r'(?:www\.)?tube\.connect\.cafe',
291 r'(?:www\.)?vid\.wxzm\.sx',
292 r'(?:www\.)?vid\.mint\.lgbt',
293 r'(?:www\.)?vid\.puffyan\.us',
294 r'(?:www\.)?yewtu\.be',
295 r'(?:www\.)?yt\.elukerio\.org',
296 r'(?:www\.)?yt\.lelux\.fi',
297 r'(?:www\.)?invidious\.ggc-project\.de',
298 r'(?:www\.)?yt\.maisputain\.ovh',
299 r'(?:www\.)?ytprivate\.com',
300 r'(?:www\.)?invidious\.13ad\.de',
301 r'(?:www\.)?invidious\.toot\.koeln',
302 r'(?:www\.)?invidious\.fdn\.fr',
303 r'(?:www\.)?watch\.nettohikari\.com',
304 r'(?:www\.)?invidious\.namazso\.eu',
305 r'(?:www\.)?invidious\.silkky\.cloud',
306 r'(?:www\.)?invidious\.exonip\.de',
307 r'(?:www\.)?invidious\.riverside\.rocks',
308 r'(?:www\.)?invidious\.blamefran\.net',
309 r'(?:www\.)?invidious\.moomoo\.de',
310 r'(?:www\.)?ytb\.trom\.tf',
311 r'(?:www\.)?yt\.cyberhost\.uk',
312 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
313 r'(?:www\.)?qklhadlycap4cnod\.onion',
314 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
315 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
316 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
317 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
318 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
319 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
320 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
321 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
322 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
323 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
324 )
325
326 def _login(self):
327 """
328 Attempt to log in to YouTube.
329 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
330 """
331
332 if (self._LOGIN_REQUIRED
333 and self.get_param('cookiefile') is None
334 and self.get_param('cookiesfrombrowser') is None):
335 self.raise_login_required(
336 'Login details are needed to download this content', method='cookies')
337 username, password = self._get_login_info()
338 if username:
339 self.report_warning(f'Cannot login to YouTube using username and password. {self._LOGIN_HINTS["cookies"]}')
340
341 def _initialize_consent(self):
342 cookies = self._get_cookies('https://www.youtube.com/')
343 if cookies.get('__Secure-3PSID'):
344 return
345 consent_id = None
346 consent = cookies.get('CONSENT')
347 if consent:
348 if 'YES' in consent.value:
349 return
350 consent_id = self._search_regex(
351 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
352 if not consent_id:
353 consent_id = random.randint(100, 999)
354 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
355
356 def _real_initialize(self):
357 self._initialize_consent()
358 self._login()
359
360 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
361 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
362 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
363
364 def _get_default_ytcfg(self, client='web'):
365 return copy.deepcopy(INNERTUBE_CLIENTS[client])
366
367 def _get_innertube_host(self, client='web'):
368 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
369
370 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
371 # try_get but with fallback to default ytcfg client values when present
372 _func = lambda y: try_get(y, getter, expected_type)
373 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
374
375 def _extract_client_name(self, ytcfg, default_client='web'):
376 return self._ytcfg_get_safe(
377 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
378 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
379
380 def _extract_client_version(self, ytcfg, default_client='web'):
381 return self._ytcfg_get_safe(
382 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
383 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
384
385 def _extract_api_key(self, ytcfg=None, default_client='web'):
386 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
387
388 def _extract_context(self, ytcfg=None, default_client='web'):
389 _get_context = lambda y: try_get(y, lambda x: x['INNERTUBE_CONTEXT'], dict)
390 context = _get_context(ytcfg)
391 if context:
392 return context
393
394 context = _get_context(self._get_default_ytcfg(default_client))
395 if not ytcfg:
396 return context
397
398 # Recreate the client context (required)
399 context['client'].update({
400 'clientVersion': self._extract_client_version(ytcfg, default_client),
401 'clientName': self._extract_client_name(ytcfg, default_client),
402 })
403 visitor_data = try_get(ytcfg, lambda x: x['VISITOR_DATA'], compat_str)
404 if visitor_data:
405 context['client']['visitorData'] = visitor_data
406 return context
407
408 _SAPISID = None
409
410 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
411 time_now = round(time.time())
412 if self._SAPISID is None:
413 yt_cookies = self._get_cookies('https://www.youtube.com')
414 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
415 # See: https://github.com/yt-dlp/yt-dlp/issues/393
416 sapisid_cookie = dict_get(
417 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
418 if sapisid_cookie and sapisid_cookie.value:
419 self._SAPISID = sapisid_cookie.value
420 self.write_debug('Extracted SAPISID cookie')
421 # SAPISID cookie is required if not already present
422 if not yt_cookies.get('SAPISID'):
423 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
424 self._set_cookie(
425 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
426 else:
427 self._SAPISID = False
428 if not self._SAPISID:
429 return None
430 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
431 sapisidhash = hashlib.sha1(
432 f'{time_now} {self._SAPISID} {origin}'.encode('utf-8')).hexdigest()
433 return f'SAPISIDHASH {time_now}_{sapisidhash}'
434
435 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
436 note='Downloading API JSON', errnote='Unable to download API page',
437 context=None, api_key=None, api_hostname=None, default_client='web'):
438
439 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
440 data.update(query)
441 real_headers = self.generate_api_headers(default_client=default_client)
442 real_headers.update({'content-type': 'application/json'})
443 if headers:
444 real_headers.update(headers)
445 return self._download_json(
446 'https://%s/youtubei/v1/%s' % (api_hostname or self._get_innertube_host(default_client), ep),
447 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
448 data=json.dumps(data).encode('utf8'), headers=real_headers,
449 query={'key': api_key or self._extract_api_key()})
450
451 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
452 data = self._search_regex(
453 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
454 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
455 if data:
456 return self._parse_json(data, item_id, fatal=fatal)
457
458 @staticmethod
459 def _extract_session_index(*data):
460 """
461 Index of current account in account list.
462 See: https://github.com/yt-dlp/yt-dlp/pull/519
463 """
464 for ytcfg in data:
465 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
466 if session_index is not None:
467 return session_index
468
469 # Deprecated?
470 def _extract_identity_token(self, ytcfg=None, webpage=None):
471 if ytcfg:
472 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
473 if token:
474 return token
475 if webpage:
476 return self._search_regex(
477 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
478 'identity token', default=None, fatal=False)
479
480 @staticmethod
481 def _extract_account_syncid(*args):
482 """
483 Extract syncId required to download private playlists of secondary channels
484 @params response and/or ytcfg
485 """
486 for data in args:
487 # ytcfg includes channel_syncid if on secondary channel
488 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
489 if delegated_sid:
490 return delegated_sid
491 sync_ids = (try_get(
492 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
493 lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
494 if len(sync_ids) >= 2 and sync_ids[1]:
495 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
496 # and just "user_syncid||" for primary channel. We only want the channel_syncid
497 return sync_ids[0]
498
499 @staticmethod
500 def _extract_visitor_data(*args):
501 """
502 Extracts visitorData from an API response or ytcfg
503 Appears to be used to track session state
504 """
505 return traverse_obj(
506 args, (..., ('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))),
507 expected_type=compat_str, get_all=False)
508
509 @property
510 def is_authenticated(self):
511 return bool(self._generate_sapisidhash_header())
512
513 def extract_ytcfg(self, video_id, webpage):
514 if not webpage:
515 return {}
516 return self._parse_json(
517 self._search_regex(
518 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
519 default='{}'), video_id, fatal=False) or {}
520
521 def generate_api_headers(
522 self, *, ytcfg=None, account_syncid=None, session_index=None,
523 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
524
525 origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client))
526 headers = {
527 'X-YouTube-Client-Name': compat_str(
528 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
529 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
530 'Origin': origin,
531 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
532 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
533 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg)
534 }
535 if session_index is None:
536 session_index = self._extract_session_index(ytcfg)
537 if account_syncid or session_index is not None:
538 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
539
540 auth = self._generate_sapisidhash_header(origin)
541 if auth is not None:
542 headers['Authorization'] = auth
543 headers['X-Origin'] = origin
544 return {h: v for h, v in headers.items() if v is not None}
545
546 @staticmethod
547 def _build_api_continuation_query(continuation, ctp=None):
548 query = {
549 'continuation': continuation
550 }
551 # TODO: Inconsistency with clickTrackingParams.
552 # Currently we have a fixed ctp contained within context (from ytcfg)
553 # and a ctp in root query for continuation.
554 if ctp:
555 query['clickTracking'] = {'clickTrackingParams': ctp}
556 return query
557
558 @classmethod
559 def _extract_next_continuation_data(cls, renderer):
560 next_continuation = try_get(
561 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
562 lambda x: x['continuation']['reloadContinuationData']), dict)
563 if not next_continuation:
564 return
565 continuation = next_continuation.get('continuation')
566 if not continuation:
567 return
568 ctp = next_continuation.get('clickTrackingParams')
569 return cls._build_api_continuation_query(continuation, ctp)
570
571 @classmethod
572 def _extract_continuation_ep_data(cls, continuation_ep: dict):
573 if isinstance(continuation_ep, dict):
574 continuation = try_get(
575 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
576 if not continuation:
577 return
578 ctp = continuation_ep.get('clickTrackingParams')
579 return cls._build_api_continuation_query(continuation, ctp)
580
581 @classmethod
582 def _extract_continuation(cls, renderer):
583 next_continuation = cls._extract_next_continuation_data(renderer)
584 if next_continuation:
585 return next_continuation
586
587 contents = []
588 for key in ('contents', 'items'):
589 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
590
591 for content in contents:
592 if not isinstance(content, dict):
593 continue
594 continuation_ep = try_get(
595 content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'],
596 lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']),
597 dict)
598 continuation = cls._extract_continuation_ep_data(continuation_ep)
599 if continuation:
600 return continuation
601
602 @classmethod
603 def _extract_alerts(cls, data):
604 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
605 if not isinstance(alert_dict, dict):
606 continue
607 for alert in alert_dict.values():
608 alert_type = alert.get('type')
609 if not alert_type:
610 continue
611 message = cls._get_text(alert, 'text')
612 if message:
613 yield alert_type, message
614
615 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
616 errors = []
617 warnings = []
618 for alert_type, alert_message in alerts:
619 if alert_type.lower() == 'error' and fatal:
620 errors.append([alert_type, alert_message])
621 else:
622 warnings.append([alert_type, alert_message])
623
624 for alert_type, alert_message in (warnings + errors[:-1]):
625 self.report_warning('YouTube said: %s - %s' % (alert_type, alert_message), only_once=only_once)
626 if errors:
627 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
628
629 def _extract_and_report_alerts(self, data, *args, **kwargs):
630 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
631
632 def _extract_badges(self, renderer: dict):
633 badges = set()
634 for badge in try_get(renderer, lambda x: x['badges'], list) or []:
635 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
636 if label:
637 badges.add(label.lower())
638 return badges
639
640 @staticmethod
641 def _get_text(data, *path_list, max_runs=None):
642 for path in path_list or [None]:
643 if path is None:
644 obj = [data]
645 else:
646 obj = traverse_obj(data, path, default=[])
647 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
648 obj = [obj]
649 for item in obj:
650 text = try_get(item, lambda x: x['simpleText'], compat_str)
651 if text:
652 return text
653 runs = try_get(item, lambda x: x['runs'], list) or []
654 if not runs and isinstance(item, list):
655 runs = item
656
657 runs = runs[:min(len(runs), max_runs or len(runs))]
658 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
659 if text:
660 return text
661
662 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
663 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
664 default_client='web'):
665 response = None
666 last_error = None
667 count = -1
668 retries = self.get_param('extractor_retries', 3)
669 if check_get_keys is None:
670 check_get_keys = []
671 while count < retries:
672 count += 1
673 if last_error:
674 self.report_warning('%s. Retrying ...' % remove_end(last_error, '.'))
675 try:
676 response = self._call_api(
677 ep=ep, fatal=True, headers=headers,
678 video_id=item_id, query=query,
679 context=self._extract_context(ytcfg, default_client),
680 api_key=self._extract_api_key(ytcfg, default_client),
681 api_hostname=api_hostname, default_client=default_client,
682 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
683 except ExtractorError as e:
684 if isinstance(e.cause, network_exceptions):
685 if isinstance(e.cause, compat_HTTPError) and not is_html(e.cause.read(512)):
686 e.cause.seek(0)
687 yt_error = try_get(
688 self._parse_json(e.cause.read().decode(), item_id, fatal=False),
689 lambda x: x['error']['message'], compat_str)
690 if yt_error:
691 self._report_alerts([('ERROR', yt_error)], fatal=False)
692 # Downloading page may result in intermittent 5xx HTTP error
693 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
694 # We also want to catch all other network exceptions since errors in later pages can be troublesome
695 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
696 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
697 last_error = error_to_compat_str(e.cause or e.msg)
698 if count < retries:
699 continue
700 if fatal:
701 raise
702 else:
703 self.report_warning(error_to_compat_str(e))
704 return
705
706 else:
707 try:
708 self._extract_and_report_alerts(response, only_once=True)
709 except ExtractorError as e:
710 # YouTube servers may return errors we want to retry on in a 200 OK response
711 # See: https://github.com/yt-dlp/yt-dlp/issues/839
712 if 'unknown error' in e.msg.lower():
713 last_error = e.msg
714 continue
715 if fatal:
716 raise
717 self.report_warning(error_to_compat_str(e))
718 return
719 if not check_get_keys or dict_get(response, check_get_keys):
720 break
721 # Youtube sometimes sends incomplete data
722 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
723 last_error = 'Incomplete data received'
724 if count >= retries:
725 if fatal:
726 raise ExtractorError(last_error)
727 else:
728 self.report_warning(last_error)
729 return
730 return response
731
732 @staticmethod
733 def is_music_url(url):
734 return re.match(r'https?://music\.youtube\.com/', url) is not None
735
736 def _extract_video(self, renderer):
737 video_id = renderer.get('videoId')
738 title = self._get_text(renderer, 'title')
739 description = self._get_text(renderer, 'descriptionSnippet')
740 duration = parse_duration(self._get_text(
741 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
742 view_count_text = self._get_text(renderer, 'viewCountText') or ''
743 view_count = str_to_int(self._search_regex(
744 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
745 'view count', default=None))
746
747 uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
748
749 return {
750 '_type': 'url',
751 'ie_key': YoutubeIE.ie_key(),
752 'id': video_id,
753 'url': f'https://www.youtube.com/watch?v={video_id}',
754 'title': title,
755 'description': description,
756 'duration': duration,
757 'view_count': view_count,
758 'uploader': uploader,
759 }
760
761
762 class YoutubeIE(YoutubeBaseInfoExtractor):
763 IE_DESC = 'YouTube'
764 _VALID_URL = r"""(?x)^
765 (
766 (?:https?://|//) # http(s):// or protocol-independent URL
767 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
768 (?:www\.)?deturl\.com/www\.youtube\.com|
769 (?:www\.)?pwnyoutube\.com|
770 (?:www\.)?hooktube\.com|
771 (?:www\.)?yourepeat\.com|
772 tube\.majestyc\.net|
773 %(invidious)s|
774 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
775 (?:.*?\#/)? # handle anchor (#/) redirect urls
776 (?: # the various things that can precede the ID:
777 (?:(?:v|embed|e|shorts)/(?!videoseries)) # v/ or embed/ or e/ or shorts/
778 |(?: # or the v= param in all its forms
779 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
780 (?:\?|\#!?) # the params delimiter ? or # or #!
781 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
782 v=
783 )
784 ))
785 |(?:
786 youtu\.be| # just youtu.be/xxxx
787 vid\.plus| # or vid.plus/xxxx
788 zwearz\.com/watch| # or zwearz.com/watch/xxxx
789 %(invidious)s
790 )/
791 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
792 )
793 )? # all until now is optional -> you can pass the naked ID
794 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
795 (?(1).+)? # if we found the ID, everything can follow
796 (?:\#|$)""" % {
797 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
798 }
799 _PLAYER_INFO_RE = (
800 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
801 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
802 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
803 )
804 _formats = {
805 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
806 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
807 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
808 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
809 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
810 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
811 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
812 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
813 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
814 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
815 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
816 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
817 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
818 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
819 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
820 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
821 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
822 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
823
824
825 # 3D videos
826 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
827 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
828 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
829 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
830 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
831 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
832 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
833
834 # Apple HTTP Live Streaming
835 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
836 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
837 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
838 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
839 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
840 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
841 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
842 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
843
844 # DASH mp4 video
845 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
846 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
847 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
848 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
849 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
850 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
851 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
852 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
853 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
854 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
855 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
856 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
857
858 # Dash mp4 audio
859 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
860 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
861 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
862 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
863 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
864 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
865 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
866
867 # Dash webm
868 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
869 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
870 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
871 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
872 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
873 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
874 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
875 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
876 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
877 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
878 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
879 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
880 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
881 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
882 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
883 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
884 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
885 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
886 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
887 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
888 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
889 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
890
891 # Dash webm audio
892 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
893 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
894
895 # Dash webm audio with opus inside
896 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
897 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
898 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
899
900 # RTMP (unnamed)
901 '_rtmp': {'protocol': 'rtmp'},
902
903 # av01 video only formats sometimes served with "unknown" codecs
904 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
905 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
906 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
907 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
908 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
909 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
910 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
911 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
912 }
913 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
914
915 _GEO_BYPASS = False
916
917 IE_NAME = 'youtube'
918 _TESTS = [
919 {
920 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
921 'info_dict': {
922 'id': 'BaW_jenozKc',
923 'ext': 'mp4',
924 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
925 'uploader': 'Philipp Hagemeister',
926 'uploader_id': 'phihag',
927 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
928 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
929 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
930 'upload_date': '20121002',
931 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
932 'categories': ['Science & Technology'],
933 'tags': ['youtube-dl'],
934 'duration': 10,
935 'view_count': int,
936 'like_count': int,
937 'dislike_count': int,
938 'start_time': 1,
939 'end_time': 9,
940 }
941 },
942 {
943 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
944 'note': 'Embed-only video (#1746)',
945 'info_dict': {
946 'id': 'yZIXLfi8CZQ',
947 'ext': 'mp4',
948 'upload_date': '20120608',
949 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
950 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
951 'uploader': 'SET India',
952 'uploader_id': 'setindia',
953 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
954 'age_limit': 18,
955 },
956 'skip': 'Private video',
957 },
958 {
959 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
960 'note': 'Use the first video ID in the URL',
961 'info_dict': {
962 'id': 'BaW_jenozKc',
963 'ext': 'mp4',
964 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
965 'uploader': 'Philipp Hagemeister',
966 'uploader_id': 'phihag',
967 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
968 'upload_date': '20121002',
969 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
970 'categories': ['Science & Technology'],
971 'tags': ['youtube-dl'],
972 'duration': 10,
973 'view_count': int,
974 'like_count': int,
975 'dislike_count': int,
976 },
977 'params': {
978 'skip_download': True,
979 },
980 },
981 {
982 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
983 'note': '256k DASH audio (format 141) via DASH manifest',
984 'info_dict': {
985 'id': 'a9LDPn-MO4I',
986 'ext': 'm4a',
987 'upload_date': '20121002',
988 'uploader_id': '8KVIDEO',
989 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
990 'description': '',
991 'uploader': '8KVIDEO',
992 'title': 'UHDTV TEST 8K VIDEO.mp4'
993 },
994 'params': {
995 'youtube_include_dash_manifest': True,
996 'format': '141',
997 },
998 'skip': 'format 141 not served anymore',
999 },
1000 # DASH manifest with encrypted signature
1001 {
1002 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1003 'info_dict': {
1004 'id': 'IB3lcPjvWLA',
1005 'ext': 'm4a',
1006 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1007 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1008 'duration': 244,
1009 'uploader': 'AfrojackVEVO',
1010 'uploader_id': 'AfrojackVEVO',
1011 'upload_date': '20131011',
1012 'abr': 129.495,
1013 },
1014 'params': {
1015 'youtube_include_dash_manifest': True,
1016 'format': '141/bestaudio[ext=m4a]',
1017 },
1018 },
1019 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1020 {
1021 'note': 'Embed allowed age-gate video',
1022 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1023 'info_dict': {
1024 'id': 'HtVdAasjOgU',
1025 'ext': 'mp4',
1026 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1027 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1028 'duration': 142,
1029 'uploader': 'The Witcher',
1030 'uploader_id': 'WitcherGame',
1031 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1032 'upload_date': '20140605',
1033 'age_limit': 18,
1034 },
1035 },
1036 {
1037 'note': 'Age-gate video with embed allowed in public site',
1038 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1039 'info_dict': {
1040 'id': 'HsUATh_Nc2U',
1041 'ext': 'mp4',
1042 'title': 'Godzilla 2 (Official Video)',
1043 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1044 'upload_date': '20200408',
1045 'uploader_id': 'FlyingKitty900',
1046 'uploader': 'FlyingKitty',
1047 'age_limit': 18,
1048 },
1049 },
1050 {
1051 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1052 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1053 'info_dict': {
1054 'id': 'Tq92D6wQ1mg',
1055 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1056 'ext': 'mp4',
1057 'upload_date': '20191227',
1058 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1059 'uploader': 'Projekt Melody',
1060 'description': 'md5:17eccca93a786d51bc67646756894066',
1061 'age_limit': 18,
1062 },
1063 },
1064 {
1065 'note': 'Non-Agegated non-embeddable video',
1066 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1067 'info_dict': {
1068 'id': 'MeJVWBSsPAY',
1069 'ext': 'mp4',
1070 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1071 'uploader': 'Herr Lurik',
1072 'uploader_id': 'st3in234',
1073 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1074 'upload_date': '20130730',
1075 },
1076 },
1077 {
1078 'note': 'Non-bypassable age-gated video',
1079 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1080 'only_matching': True,
1081 },
1082 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1083 # YouTube Red ad is not captured for creator
1084 {
1085 'url': '__2ABJjxzNo',
1086 'info_dict': {
1087 'id': '__2ABJjxzNo',
1088 'ext': 'mp4',
1089 'duration': 266,
1090 'upload_date': '20100430',
1091 'uploader_id': 'deadmau5',
1092 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1093 'creator': 'deadmau5',
1094 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1095 'uploader': 'deadmau5',
1096 'title': 'Deadmau5 - Some Chords (HD)',
1097 'alt_title': 'Some Chords',
1098 },
1099 'expected_warnings': [
1100 'DASH manifest missing',
1101 ]
1102 },
1103 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1104 {
1105 'url': 'lqQg6PlCWgI',
1106 'info_dict': {
1107 'id': 'lqQg6PlCWgI',
1108 'ext': 'mp4',
1109 'duration': 6085,
1110 'upload_date': '20150827',
1111 'uploader_id': 'olympic',
1112 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1113 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
1114 'uploader': 'Olympics',
1115 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1116 },
1117 'params': {
1118 'skip_download': 'requires avconv',
1119 }
1120 },
1121 # Non-square pixels
1122 {
1123 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1124 'info_dict': {
1125 'id': '_b-2C3KPAM0',
1126 'ext': 'mp4',
1127 'stretched_ratio': 16 / 9.,
1128 'duration': 85,
1129 'upload_date': '20110310',
1130 'uploader_id': 'AllenMeow',
1131 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1132 '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',
1133 'uploader': '孫ᄋᄅ',
1134 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1135 },
1136 },
1137 # url_encoded_fmt_stream_map is empty string
1138 {
1139 'url': 'qEJwOuvDf7I',
1140 'info_dict': {
1141 'id': 'qEJwOuvDf7I',
1142 'ext': 'webm',
1143 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1144 'description': '',
1145 'upload_date': '20150404',
1146 'uploader_id': 'spbelect',
1147 'uploader': 'Наблюдатели Петербурга',
1148 },
1149 'params': {
1150 'skip_download': 'requires avconv',
1151 },
1152 'skip': 'This live event has ended.',
1153 },
1154 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1155 {
1156 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1157 'info_dict': {
1158 'id': 'FIl7x6_3R5Y',
1159 'ext': 'webm',
1160 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1161 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1162 'duration': 220,
1163 'upload_date': '20150625',
1164 'uploader_id': 'dorappi2000',
1165 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1166 'uploader': 'dorappi2000',
1167 'formats': 'mincount:31',
1168 },
1169 'skip': 'not actual anymore',
1170 },
1171 # DASH manifest with segment_list
1172 {
1173 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1174 'md5': '8ce563a1d667b599d21064e982ab9e31',
1175 'info_dict': {
1176 'id': 'CsmdDsKjzN8',
1177 'ext': 'mp4',
1178 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1179 'uploader': 'Airtek',
1180 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1181 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1182 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1183 },
1184 'params': {
1185 'youtube_include_dash_manifest': True,
1186 'format': '135', # bestvideo
1187 },
1188 'skip': 'This live event has ended.',
1189 },
1190 {
1191 # Multifeed videos (multiple cameras), URL is for Main Camera
1192 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
1193 'info_dict': {
1194 'id': 'jvGDaLqkpTg',
1195 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
1196 'description': 'md5:e03b909557865076822aa169218d6a5d',
1197 },
1198 'playlist': [{
1199 'info_dict': {
1200 'id': 'jvGDaLqkpTg',
1201 'ext': 'mp4',
1202 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
1203 'description': 'md5:e03b909557865076822aa169218d6a5d',
1204 'duration': 10643,
1205 'upload_date': '20161111',
1206 'uploader': 'Team PGP',
1207 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1208 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1209 },
1210 }, {
1211 'info_dict': {
1212 'id': '3AKt1R1aDnw',
1213 'ext': 'mp4',
1214 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
1215 'description': 'md5:e03b909557865076822aa169218d6a5d',
1216 'duration': 10991,
1217 'upload_date': '20161111',
1218 'uploader': 'Team PGP',
1219 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1220 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1221 },
1222 }, {
1223 'info_dict': {
1224 'id': 'RtAMM00gpVc',
1225 'ext': 'mp4',
1226 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
1227 'description': 'md5:e03b909557865076822aa169218d6a5d',
1228 'duration': 10995,
1229 'upload_date': '20161111',
1230 'uploader': 'Team PGP',
1231 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1232 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1233 },
1234 }, {
1235 'info_dict': {
1236 'id': '6N2fdlP3C5U',
1237 'ext': 'mp4',
1238 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
1239 'description': 'md5:e03b909557865076822aa169218d6a5d',
1240 'duration': 10990,
1241 'upload_date': '20161111',
1242 'uploader': 'Team PGP',
1243 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
1244 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
1245 },
1246 }],
1247 'params': {
1248 'skip_download': True,
1249 },
1250 'skip': 'Not multifeed anymore',
1251 },
1252 {
1253 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1254 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1255 'info_dict': {
1256 'id': 'gVfLd0zydlo',
1257 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1258 },
1259 'playlist_count': 2,
1260 'skip': 'Not multifeed anymore',
1261 },
1262 {
1263 'url': 'https://vid.plus/FlRa-iH7PGw',
1264 'only_matching': True,
1265 },
1266 {
1267 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1268 'only_matching': True,
1269 },
1270 {
1271 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1272 # Also tests cut-off URL expansion in video description (see
1273 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1274 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1275 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1276 'info_dict': {
1277 'id': 'lsguqyKfVQg',
1278 'ext': 'mp4',
1279 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1280 'alt_title': 'Dark Walk',
1281 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1282 'duration': 133,
1283 'upload_date': '20151119',
1284 'uploader_id': 'IronSoulElf',
1285 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1286 'uploader': 'IronSoulElf',
1287 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1288 'track': 'Dark Walk',
1289 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1290 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1291 },
1292 'params': {
1293 'skip_download': True,
1294 },
1295 },
1296 {
1297 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1298 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1299 'only_matching': True,
1300 },
1301 {
1302 # Video with yt:stretch=17:0
1303 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1304 'info_dict': {
1305 'id': 'Q39EVAstoRM',
1306 'ext': 'mp4',
1307 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1308 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1309 'upload_date': '20151107',
1310 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1311 'uploader': 'CH GAMER DROID',
1312 },
1313 'params': {
1314 'skip_download': True,
1315 },
1316 'skip': 'This video does not exist.',
1317 },
1318 {
1319 # Video with incomplete 'yt:stretch=16:'
1320 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1321 'only_matching': True,
1322 },
1323 {
1324 # Video licensed under Creative Commons
1325 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1326 'info_dict': {
1327 'id': 'M4gD1WSo5mA',
1328 'ext': 'mp4',
1329 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1330 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1331 'duration': 721,
1332 'upload_date': '20150127',
1333 'uploader_id': 'BerkmanCenter',
1334 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1335 'uploader': 'The Berkman Klein Center for Internet & Society',
1336 'license': 'Creative Commons Attribution license (reuse allowed)',
1337 },
1338 'params': {
1339 'skip_download': True,
1340 },
1341 },
1342 {
1343 # Channel-like uploader_url
1344 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1345 'info_dict': {
1346 'id': 'eQcmzGIKrzg',
1347 'ext': 'mp4',
1348 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1349 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1350 'duration': 4060,
1351 'upload_date': '20151119',
1352 'uploader': 'Bernie Sanders',
1353 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1354 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1355 'license': 'Creative Commons Attribution license (reuse allowed)',
1356 },
1357 'params': {
1358 'skip_download': True,
1359 },
1360 },
1361 {
1362 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1363 'only_matching': True,
1364 },
1365 {
1366 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1367 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1368 'only_matching': True,
1369 },
1370 {
1371 # Rental video preview
1372 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1373 'info_dict': {
1374 'id': 'uGpuVWrhIzE',
1375 'ext': 'mp4',
1376 'title': 'Piku - Trailer',
1377 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1378 'upload_date': '20150811',
1379 'uploader': 'FlixMatrix',
1380 'uploader_id': 'FlixMatrixKaravan',
1381 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1382 'license': 'Standard YouTube License',
1383 },
1384 'params': {
1385 'skip_download': True,
1386 },
1387 'skip': 'This video is not available.',
1388 },
1389 {
1390 # YouTube Red video with episode data
1391 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1392 'info_dict': {
1393 'id': 'iqKdEhx-dD4',
1394 'ext': 'mp4',
1395 'title': 'Isolation - Mind Field (Ep 1)',
1396 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1397 'duration': 2085,
1398 'upload_date': '20170118',
1399 'uploader': 'Vsauce',
1400 'uploader_id': 'Vsauce',
1401 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1402 'series': 'Mind Field',
1403 'season_number': 1,
1404 'episode_number': 1,
1405 },
1406 'params': {
1407 'skip_download': True,
1408 },
1409 'expected_warnings': [
1410 'Skipping DASH manifest',
1411 ],
1412 },
1413 {
1414 # The following content has been identified by the YouTube community
1415 # as inappropriate or offensive to some audiences.
1416 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1417 'info_dict': {
1418 'id': '6SJNVb0GnPI',
1419 'ext': 'mp4',
1420 'title': 'Race Differences in Intelligence',
1421 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1422 'duration': 965,
1423 'upload_date': '20140124',
1424 'uploader': 'New Century Foundation',
1425 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1426 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1427 },
1428 'params': {
1429 'skip_download': True,
1430 },
1431 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1432 },
1433 {
1434 # itag 212
1435 'url': '1t24XAntNCY',
1436 'only_matching': True,
1437 },
1438 {
1439 # geo restricted to JP
1440 'url': 'sJL6WA-aGkQ',
1441 'only_matching': True,
1442 },
1443 {
1444 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1445 'only_matching': True,
1446 },
1447 {
1448 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1449 'only_matching': True,
1450 },
1451 {
1452 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1453 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1454 'only_matching': True,
1455 },
1456 {
1457 # DRM protected
1458 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1459 'only_matching': True,
1460 },
1461 {
1462 # Video with unsupported adaptive stream type formats
1463 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1464 'info_dict': {
1465 'id': 'Z4Vy8R84T1U',
1466 'ext': 'mp4',
1467 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1468 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1469 'duration': 433,
1470 'upload_date': '20130923',
1471 'uploader': 'Amelia Putri Harwita',
1472 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1473 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1474 'formats': 'maxcount:10',
1475 },
1476 'params': {
1477 'skip_download': True,
1478 'youtube_include_dash_manifest': False,
1479 },
1480 'skip': 'not actual anymore',
1481 },
1482 {
1483 # Youtube Music Auto-generated description
1484 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1485 'info_dict': {
1486 'id': 'MgNrAu2pzNs',
1487 'ext': 'mp4',
1488 'title': 'Voyeur Girl',
1489 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1490 'upload_date': '20190312',
1491 'uploader': 'Stephen - Topic',
1492 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1493 'artist': 'Stephen',
1494 'track': 'Voyeur Girl',
1495 'album': 'it\'s too much love to know my dear',
1496 'release_date': '20190313',
1497 'release_year': 2019,
1498 },
1499 'params': {
1500 'skip_download': True,
1501 },
1502 },
1503 {
1504 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1505 'only_matching': True,
1506 },
1507 {
1508 # invalid -> valid video id redirection
1509 'url': 'DJztXj2GPfl',
1510 'info_dict': {
1511 'id': 'DJztXj2GPfk',
1512 'ext': 'mp4',
1513 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1514 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1515 'upload_date': '20090125',
1516 'uploader': 'Prochorowka',
1517 'uploader_id': 'Prochorowka',
1518 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1519 'artist': 'Panjabi MC',
1520 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1521 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1522 },
1523 'params': {
1524 'skip_download': True,
1525 },
1526 'skip': 'Video unavailable',
1527 },
1528 {
1529 # empty description results in an empty string
1530 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1531 'info_dict': {
1532 'id': 'x41yOUIvK2k',
1533 'ext': 'mp4',
1534 'title': 'IMG 3456',
1535 'description': '',
1536 'upload_date': '20170613',
1537 'uploader_id': 'ElevageOrVert',
1538 'uploader': 'ElevageOrVert',
1539 },
1540 'params': {
1541 'skip_download': True,
1542 },
1543 },
1544 {
1545 # with '};' inside yt initial data (see [1])
1546 # see [2] for an example with '};' inside ytInitialPlayerResponse
1547 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1548 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1549 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1550 'info_dict': {
1551 'id': 'CHqg6qOn4no',
1552 'ext': 'mp4',
1553 'title': 'Part 77 Sort a list of simple types in c#',
1554 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1555 'upload_date': '20130831',
1556 'uploader_id': 'kudvenkat',
1557 'uploader': 'kudvenkat',
1558 },
1559 'params': {
1560 'skip_download': True,
1561 },
1562 },
1563 {
1564 # another example of '};' in ytInitialData
1565 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1566 'only_matching': True,
1567 },
1568 {
1569 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1570 'only_matching': True,
1571 },
1572 {
1573 # https://github.com/ytdl-org/youtube-dl/pull/28094
1574 'url': 'OtqTfy26tG0',
1575 'info_dict': {
1576 'id': 'OtqTfy26tG0',
1577 'ext': 'mp4',
1578 'title': 'Burn Out',
1579 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1580 'upload_date': '20141120',
1581 'uploader': 'The Cinematic Orchestra - Topic',
1582 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1583 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1584 'artist': 'The Cinematic Orchestra',
1585 'track': 'Burn Out',
1586 'album': 'Every Day',
1587 'release_data': None,
1588 'release_year': None,
1589 },
1590 'params': {
1591 'skip_download': True,
1592 },
1593 },
1594 {
1595 # controversial video, only works with bpctr when authenticated with cookies
1596 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1597 'only_matching': True,
1598 },
1599 {
1600 # controversial video, requires bpctr/contentCheckOk
1601 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
1602 'info_dict': {
1603 'id': 'SZJvDhaSDnc',
1604 'ext': 'mp4',
1605 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
1606 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
1607 'uploader': 'CBS This Morning',
1608 'uploader_id': 'CBSThisMorning',
1609 'upload_date': '20140716',
1610 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7'
1611 }
1612 },
1613 {
1614 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
1615 'url': 'cBvYw8_A0vQ',
1616 'info_dict': {
1617 'id': 'cBvYw8_A0vQ',
1618 'ext': 'mp4',
1619 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
1620 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
1621 'upload_date': '20201120',
1622 'uploader': 'Walk around Japan',
1623 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
1624 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
1625 },
1626 'params': {
1627 'skip_download': True,
1628 },
1629 }, {
1630 # Has multiple audio streams
1631 'url': 'WaOKSUlf4TM',
1632 'only_matching': True
1633 }, {
1634 # Requires Premium: has format 141 when requested using YTM url
1635 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
1636 'only_matching': True
1637 }, {
1638 # multiple subtitles with same lang_code
1639 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
1640 'only_matching': True,
1641 }, {
1642 # Force use android client fallback
1643 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
1644 'info_dict': {
1645 'id': 'YOelRv7fMxY',
1646 'title': 'DIGGING A SECRET TUNNEL Part 1',
1647 'ext': '3gp',
1648 'upload_date': '20210624',
1649 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
1650 'uploader': 'colinfurze',
1651 'uploader_id': 'colinfurze',
1652 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
1653 'description': 'md5:b5096f56af7ccd7a555c84db81738b22'
1654 },
1655 'params': {
1656 'format': '17', # 3gp format available on android
1657 'extractor_args': {'youtube': {'player_client': ['android']}},
1658 },
1659 },
1660 {
1661 # Skip download of additional client configs (remix client config in this case)
1662 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1663 'only_matching': True,
1664 'params': {
1665 'extractor_args': {'youtube': {'player_skip': ['configs']}},
1666 },
1667 }, {
1668 # shorts
1669 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
1670 'only_matching': True,
1671 },
1672 ]
1673
1674 @classmethod
1675 def suitable(cls, url):
1676 from ..utils import parse_qs
1677
1678 qs = parse_qs(url)
1679 if qs.get('list', [None])[0]:
1680 return False
1681 return super(YoutubeIE, cls).suitable(url)
1682
1683 def __init__(self, *args, **kwargs):
1684 super(YoutubeIE, self).__init__(*args, **kwargs)
1685 self._code_cache = {}
1686 self._player_cache = {}
1687
1688 def _extract_player_url(self, *ytcfgs, webpage=None):
1689 player_url = traverse_obj(
1690 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
1691 get_all=False, expected_type=compat_str)
1692 if not player_url:
1693 return
1694 if player_url.startswith('//'):
1695 player_url = 'https:' + player_url
1696 elif not re.match(r'https?://', player_url):
1697 player_url = compat_urlparse.urljoin(
1698 'https://www.youtube.com', player_url)
1699 return player_url
1700
1701 def _download_player_url(self, video_id, fatal=False):
1702 res = self._download_webpage(
1703 'https://www.youtube.com/iframe_api',
1704 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
1705 if res:
1706 player_version = self._search_regex(
1707 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
1708 if player_version:
1709 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
1710
1711 def _signature_cache_id(self, example_sig):
1712 """ Return a string representation of a signature """
1713 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1714
1715 @classmethod
1716 def _extract_player_info(cls, player_url):
1717 for player_re in cls._PLAYER_INFO_RE:
1718 id_m = re.search(player_re, player_url)
1719 if id_m:
1720 break
1721 else:
1722 raise ExtractorError('Cannot identify player %r' % player_url)
1723 return id_m.group('id')
1724
1725 def _load_player(self, video_id, player_url, fatal=True):
1726 player_id = self._extract_player_info(player_url)
1727 if player_id not in self._code_cache:
1728 code = self._download_webpage(
1729 player_url, video_id, fatal=fatal,
1730 note='Downloading player ' + player_id,
1731 errnote='Download of %s failed' % player_url)
1732 if code:
1733 self._code_cache[player_id] = code
1734 return self._code_cache.get(player_id)
1735
1736 def _extract_signature_function(self, video_id, player_url, example_sig):
1737 player_id = self._extract_player_info(player_url)
1738
1739 # Read from filesystem cache
1740 func_id = 'js_%s_%s' % (
1741 player_id, self._signature_cache_id(example_sig))
1742 assert os.path.basename(func_id) == func_id
1743
1744 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1745 if cache_spec is not None:
1746 return lambda s: ''.join(s[i] for i in cache_spec)
1747
1748 code = self._load_player(video_id, player_url)
1749 if code:
1750 res = self._parse_sig_js(code)
1751
1752 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1753 cache_res = res(test_string)
1754 cache_spec = [ord(c) for c in cache_res]
1755
1756 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1757 return res
1758
1759 def _print_sig_code(self, func, example_sig):
1760 if not self.get_param('youtube_print_sig_code'):
1761 return
1762
1763 def gen_sig_code(idxs):
1764 def _genslice(start, end, step):
1765 starts = '' if start == 0 else str(start)
1766 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1767 steps = '' if step == 1 else (':%d' % step)
1768 return 's[%s%s%s]' % (starts, ends, steps)
1769
1770 step = None
1771 # Quelch pyflakes warnings - start will be set when step is set
1772 start = '(Never used)'
1773 for i, prev in zip(idxs[1:], idxs[:-1]):
1774 if step is not None:
1775 if i - prev == step:
1776 continue
1777 yield _genslice(start, prev, step)
1778 step = None
1779 continue
1780 if i - prev in [-1, 1]:
1781 step = i - prev
1782 start = prev
1783 continue
1784 else:
1785 yield 's[%d]' % prev
1786 if step is None:
1787 yield 's[%d]' % i
1788 else:
1789 yield _genslice(start, i, step)
1790
1791 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1792 cache_res = func(test_string)
1793 cache_spec = [ord(c) for c in cache_res]
1794 expr_code = ' + '.join(gen_sig_code(cache_spec))
1795 signature_id_tuple = '(%s)' % (
1796 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1797 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1798 ' return %s\n') % (signature_id_tuple, expr_code)
1799 self.to_screen('Extracted signature function:\n' + code)
1800
1801 def _parse_sig_js(self, jscode):
1802 funcname = self._search_regex(
1803 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1804 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1805 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
1806 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
1807 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}\(a,\d+\)',
1808 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\(\s*""\s*\)',
1809 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1810 # Obsolete patterns
1811 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1812 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1813 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1814 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1815 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1816 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1817 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1818 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1819 jscode, 'Initial JS player signature function name', group='sig')
1820
1821 jsi = JSInterpreter(jscode)
1822 initial_function = jsi.extract_function(funcname)
1823 return lambda s: initial_function([s])
1824
1825 def _decrypt_signature(self, s, video_id, player_url):
1826 """Turn the encrypted s field into a working signature"""
1827
1828 if player_url is None:
1829 raise ExtractorError('Cannot decrypt signature without player_url')
1830
1831 try:
1832 player_id = (player_url, self._signature_cache_id(s))
1833 if player_id not in self._player_cache:
1834 func = self._extract_signature_function(
1835 video_id, player_url, s
1836 )
1837 self._player_cache[player_id] = func
1838 func = self._player_cache[player_id]
1839 self._print_sig_code(func, s)
1840 return func(s)
1841 except Exception as e:
1842 raise ExtractorError('Signature extraction failed: ' + traceback.format_exc(), cause=e)
1843
1844 def _decrypt_nsig(self, s, video_id, player_url):
1845 """Turn the encrypted n field into a working signature"""
1846 if player_url is None:
1847 raise ExtractorError('Cannot decrypt nsig without player_url')
1848 if player_url.startswith('//'):
1849 player_url = 'https:' + player_url
1850 elif not re.match(r'https?://', player_url):
1851 player_url = compat_urlparse.urljoin(
1852 'https://www.youtube.com', player_url)
1853
1854 sig_id = ('nsig_value', s)
1855 if sig_id in self._player_cache:
1856 return self._player_cache[sig_id]
1857
1858 try:
1859 player_id = ('nsig', player_url)
1860 if player_id not in self._player_cache:
1861 self._player_cache[player_id] = self._extract_n_function(video_id, player_url)
1862 func = self._player_cache[player_id]
1863 self._player_cache[sig_id] = func(s)
1864 self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}')
1865 return self._player_cache[sig_id]
1866 except Exception as e:
1867 raise ExtractorError(traceback.format_exc(), cause=e)
1868
1869 def _extract_n_function_name(self, jscode):
1870 return self._search_regex(
1871 (r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]{3})\([a-zA-Z0-9]\)',),
1872 jscode, 'Initial JS player n function name', group='nfunc')
1873
1874 def _extract_n_function(self, video_id, player_url):
1875 player_id = self._extract_player_info(player_url)
1876 func_code = self._downloader.cache.load('youtube-nsig', player_id)
1877
1878 if func_code:
1879 jsi = JSInterpreter(func_code)
1880 else:
1881 jscode = self._load_player(video_id, player_url)
1882 funcname = self._extract_n_function_name(jscode)
1883 jsi = JSInterpreter(jscode)
1884 func_code = jsi.extract_function_code(funcname)
1885 self._downloader.cache.store('youtube-nsig', player_id, func_code)
1886
1887 if self.get_param('youtube_print_sig_code'):
1888 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
1889
1890 return lambda s: jsi.extract_function_from_code(*func_code)([s])
1891
1892 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
1893 """
1894 Extract signatureTimestamp (sts)
1895 Required to tell API what sig/player version is in use.
1896 """
1897 sts = None
1898 if isinstance(ytcfg, dict):
1899 sts = int_or_none(ytcfg.get('STS'))
1900
1901 if not sts:
1902 # Attempt to extract from player
1903 if player_url is None:
1904 error_msg = 'Cannot extract signature timestamp without player_url.'
1905 if fatal:
1906 raise ExtractorError(error_msg)
1907 self.report_warning(error_msg)
1908 return
1909 code = self._load_player(video_id, player_url, fatal=fatal)
1910 if code:
1911 sts = int_or_none(self._search_regex(
1912 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
1913 'JS player signature timestamp', group='sts', fatal=fatal))
1914 return sts
1915
1916 def _mark_watched(self, video_id, player_responses):
1917 playback_url = traverse_obj(
1918 player_responses, (..., 'playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'),
1919 expected_type=url_or_none, get_all=False)
1920 if not playback_url:
1921 self.report_warning('Unable to mark watched')
1922 return
1923 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1924 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1925
1926 # cpn generation algorithm is reverse engineered from base.js.
1927 # In fact it works even with dummy cpn.
1928 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1929 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1930
1931 qs.update({
1932 'ver': ['2'],
1933 'cpn': [cpn],
1934 })
1935 playback_url = compat_urlparse.urlunparse(
1936 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1937
1938 self._download_webpage(
1939 playback_url, video_id, 'Marking watched',
1940 'Unable to mark watched', fatal=False)
1941
1942 @staticmethod
1943 def _extract_urls(webpage):
1944 # Embedded YouTube player
1945 entries = [
1946 unescapeHTML(mobj.group('url'))
1947 for mobj in re.finditer(r'''(?x)
1948 (?:
1949 <iframe[^>]+?src=|
1950 data-video-url=|
1951 <embed[^>]+?src=|
1952 embedSWF\(?:\s*|
1953 <object[^>]+data=|
1954 new\s+SWFObject\(
1955 )
1956 (["\'])
1957 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1958 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1959 \1''', webpage)]
1960
1961 # lazyYT YouTube embed
1962 entries.extend(list(map(
1963 unescapeHTML,
1964 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1965
1966 # Wordpress "YouTube Video Importer" plugin
1967 matches = re.findall(r'''(?x)<div[^>]+
1968 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1969 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1970 entries.extend(m[-1] for m in matches)
1971
1972 return entries
1973
1974 @staticmethod
1975 def _extract_url(webpage):
1976 urls = YoutubeIE._extract_urls(webpage)
1977 return urls[0] if urls else None
1978
1979 @classmethod
1980 def extract_id(cls, url):
1981 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1982 if mobj is None:
1983 raise ExtractorError('Invalid URL: %s' % url)
1984 return mobj.group('id')
1985
1986 def _extract_chapters_from_json(self, data, duration):
1987 chapter_list = traverse_obj(
1988 data, (
1989 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
1990 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
1991 ), expected_type=list)
1992
1993 return self._extract_chapters(
1994 chapter_list,
1995 chapter_time=lambda chapter: float_or_none(
1996 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
1997 chapter_title=lambda chapter: traverse_obj(
1998 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
1999 duration=duration)
2000
2001 def _extract_chapters_from_engagement_panel(self, data, duration):
2002 content_list = traverse_obj(
2003 data,
2004 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
2005 expected_type=list, default=[])
2006 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
2007 chapter_title = lambda chapter: self._get_text(chapter, 'title')
2008
2009 return next((
2010 filter(None, (
2011 self._extract_chapters(
2012 traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
2013 chapter_time, chapter_title, duration)
2014 for contents in content_list
2015 ))), [])
2016
2017 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration):
2018 chapters = []
2019 last_chapter = {'start_time': 0}
2020 for idx, chapter in enumerate(chapter_list or []):
2021 title = chapter_title(chapter)
2022 start_time = chapter_time(chapter)
2023 if start_time is None:
2024 continue
2025 last_chapter['end_time'] = start_time
2026 if start_time < last_chapter['start_time']:
2027 if idx == 1:
2028 chapters.pop()
2029 self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title'])
2030 else:
2031 self.report_warning(f'Invalid start time for chapter "{title}"')
2032 continue
2033 last_chapter = {'start_time': start_time, 'title': title}
2034 chapters.append(last_chapter)
2035 last_chapter['end_time'] = duration
2036 return chapters
2037
2038 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
2039 return self._parse_json(self._search_regex(
2040 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
2041 regex), webpage, name, default='{}'), video_id, fatal=False)
2042
2043 @staticmethod
2044 def parse_time_text(time_text):
2045 """
2046 Parse the comment time text
2047 time_text is in the format 'X units ago (edited)'
2048 """
2049 time_text_split = time_text.split(' ')
2050 if len(time_text_split) >= 3:
2051 try:
2052 return datetime_from_str('now-%s%s' % (time_text_split[0], time_text_split[1]), precision='auto')
2053 except ValueError:
2054 return None
2055
2056 def _extract_comment(self, comment_renderer, parent=None):
2057 comment_id = comment_renderer.get('commentId')
2058 if not comment_id:
2059 return
2060
2061 text = self._get_text(comment_renderer, 'contentText')
2062
2063 # note: timestamp is an estimate calculated from the current time and time_text
2064 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
2065 time_text_dt = self.parse_time_text(time_text)
2066 if isinstance(time_text_dt, datetime.datetime):
2067 timestamp = calendar.timegm(time_text_dt.timetuple())
2068 author = self._get_text(comment_renderer, 'authorText')
2069 author_id = try_get(comment_renderer,
2070 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
2071
2072 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
2073 lambda x: x['likeCount']), compat_str)) or 0
2074 author_thumbnail = try_get(comment_renderer,
2075 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
2076
2077 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
2078 is_favorited = 'creatorHeart' in (try_get(
2079 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
2080 return {
2081 'id': comment_id,
2082 'text': text,
2083 'timestamp': timestamp,
2084 'time_text': time_text,
2085 'like_count': votes,
2086 'is_favorited': is_favorited,
2087 'author': author,
2088 'author_id': author_id,
2089 'author_thumbnail': author_thumbnail,
2090 'author_is_uploader': author_is_uploader,
2091 'parent': parent or 'root'
2092 }
2093
2094 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, comment_counts=None):
2095
2096 def extract_header(contents):
2097 _continuation = None
2098 for content in contents:
2099 comments_header_renderer = try_get(content, lambda x: x['commentsHeaderRenderer'])
2100 expected_comment_count = parse_count(self._get_text(
2101 comments_header_renderer, 'countText', 'commentsCount', max_runs=1))
2102
2103 if expected_comment_count:
2104 comment_counts[1] = expected_comment_count
2105 self.to_screen('Downloading ~%d comments' % expected_comment_count)
2106 sort_mode_str = self._configuration_arg('comment_sort', [''])[0]
2107 comment_sort_index = int(sort_mode_str != 'top') # 1 = new, 0 = top
2108
2109 sort_menu_item = try_get(
2110 comments_header_renderer,
2111 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
2112 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
2113
2114 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
2115 if not _continuation:
2116 continue
2117
2118 sort_text = sort_menu_item.get('title')
2119 if isinstance(sort_text, compat_str):
2120 sort_text = sort_text.lower()
2121 else:
2122 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
2123 self.to_screen('Sorting comments by %s' % sort_text)
2124 break
2125 return _continuation
2126
2127 def extract_thread(contents):
2128 if not parent:
2129 comment_counts[2] = 0
2130 for content in contents:
2131 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
2132 comment_renderer = try_get(
2133 comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
2134 content, (lambda x: x['commentRenderer'], dict))
2135
2136 if not comment_renderer:
2137 continue
2138 comment = self._extract_comment(comment_renderer, parent)
2139 if not comment:
2140 continue
2141 comment_counts[0] += 1
2142 yield comment
2143 # Attempt to get the replies
2144 comment_replies_renderer = try_get(
2145 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
2146
2147 if comment_replies_renderer:
2148 comment_counts[2] += 1
2149 comment_entries_iter = self._comment_entries(
2150 comment_replies_renderer, ytcfg, video_id,
2151 parent=comment.get('id'), comment_counts=comment_counts)
2152
2153 for reply_comment in comment_entries_iter:
2154 yield reply_comment
2155
2156 # YouTube comments have a max depth of 2
2157 max_depth = int_or_none(self._configuration_arg('max_comment_depth', [''])[0]) or float('inf')
2158 if max_depth == 1 and parent:
2159 return
2160 if not comment_counts:
2161 # comment so far, est. total comments, current comment thread #
2162 comment_counts = [0, 0, 0]
2163
2164 continuation = self._extract_continuation(root_continuation_data)
2165 if continuation and len(continuation['continuation']) < 27:
2166 self.write_debug('Detected old API continuation token. Generating new API compatible token.')
2167 continuation_token = self._generate_comment_continuation(video_id)
2168 continuation = self._build_api_continuation_query(continuation_token, None)
2169
2170 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
2171 if message and not parent:
2172 self.report_warning(message, video_id=video_id)
2173
2174 visitor_data = None
2175 is_first_continuation = parent is None
2176
2177 for page_num in itertools.count(0):
2178 if not continuation:
2179 break
2180 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=visitor_data)
2181 comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
2182 if page_num == 0:
2183 if is_first_continuation:
2184 note_prefix = 'Downloading comment section API JSON'
2185 else:
2186 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
2187 comment_counts[2], comment_prog_str)
2188 else:
2189 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
2190 ' ' if parent else '', ' replies' if parent else '',
2191 page_num, comment_prog_str)
2192
2193 response = self._extract_response(
2194 item_id=None, query=continuation,
2195 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
2196 check_get_keys=('onResponseReceivedEndpoints', 'continuationContents'))
2197 if not response:
2198 break
2199 visitor_data = try_get(
2200 response,
2201 lambda x: x['responseContext']['webResponseContextExtensionData']['ytConfigData']['visitorData'],
2202 compat_str) or visitor_data
2203
2204 continuation_contents = dict_get(response, ('onResponseReceivedEndpoints', 'continuationContents'))
2205
2206 continuation = None
2207 if isinstance(continuation_contents, list):
2208 for continuation_section in continuation_contents:
2209 if not isinstance(continuation_section, dict):
2210 continue
2211 continuation_items = try_get(
2212 continuation_section,
2213 (lambda x: x['reloadContinuationItemsCommand']['continuationItems'],
2214 lambda x: x['appendContinuationItemsAction']['continuationItems']),
2215 list) or []
2216 if is_first_continuation:
2217 continuation = extract_header(continuation_items)
2218 is_first_continuation = False
2219 if continuation:
2220 break
2221 continue
2222 count = 0
2223 for count, entry in enumerate(extract_thread(continuation_items)):
2224 yield entry
2225 continuation = self._extract_continuation({'contents': continuation_items})
2226 if continuation:
2227 # Sometimes YouTube provides a continuation without any comments
2228 # In most cases we end up just downloading these with very little comments to come.
2229 if count == 0:
2230 if not parent:
2231 self.report_warning('No comments received - assuming end of comments')
2232 continuation = None
2233 break
2234
2235 # Deprecated response structure
2236 elif isinstance(continuation_contents, dict):
2237 known_continuation_renderers = ('itemSectionContinuation', 'commentRepliesContinuation')
2238 for key, continuation_renderer in continuation_contents.items():
2239 if key not in known_continuation_renderers:
2240 continue
2241 if not isinstance(continuation_renderer, dict):
2242 continue
2243 if is_first_continuation:
2244 header_continuation_items = [continuation_renderer.get('header') or {}]
2245 continuation = extract_header(header_continuation_items)
2246 is_first_continuation = False
2247 if continuation:
2248 break
2249
2250 # Sometimes YouTube provides a continuation without any comments
2251 # In most cases we end up just downloading these with very little comments to come.
2252 count = 0
2253 for count, entry in enumerate(extract_thread(continuation_renderer.get('contents') or {})):
2254 yield entry
2255 continuation = self._extract_continuation(continuation_renderer)
2256 if count == 0:
2257 if not parent:
2258 self.report_warning('No comments received - assuming end of comments')
2259 continuation = None
2260 break
2261
2262 @staticmethod
2263 def _generate_comment_continuation(video_id):
2264 """
2265 Generates initial comment section continuation token from given video id
2266 """
2267 b64_vid_id = base64.b64encode(bytes(video_id.encode('utf-8')))
2268 parts = ('Eg0SCw==', b64_vid_id, 'GAYyJyIRIgs=', b64_vid_id, 'MAB4AjAAQhBjb21tZW50cy1zZWN0aW9u')
2269 new_continuation_intlist = list(itertools.chain.from_iterable(
2270 [bytes_to_intlist(base64.b64decode(part)) for part in parts]))
2271 return base64.b64encode(intlist_to_bytes(new_continuation_intlist)).decode('utf-8')
2272
2273 def _get_comments(self, ytcfg, video_id, contents, webpage):
2274 """Entry for comment extraction"""
2275 def _real_comment_extract(contents):
2276 renderer = next((
2277 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
2278 if item.get('sectionIdentifier') == 'comment-item-section'), None)
2279 yield from self._comment_entries(renderer, ytcfg, video_id)
2280
2281 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
2282 # Force English regardless of account setting to prevent parsing issues
2283 # See: https://github.com/yt-dlp/yt-dlp/issues/532
2284 ytcfg = copy.deepcopy(ytcfg)
2285 traverse_obj(
2286 ytcfg, ('INNERTUBE_CONTEXT', 'client'), expected_type=dict, default={})['hl'] = 'en'
2287 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
2288
2289 @staticmethod
2290 def _get_checkok_params():
2291 return {'contentCheckOk': True, 'racyCheckOk': True}
2292
2293 @classmethod
2294 def _generate_player_context(cls, sts=None):
2295 context = {
2296 'html5Preference': 'HTML5_PREF_WANTS',
2297 }
2298 if sts is not None:
2299 context['signatureTimestamp'] = sts
2300 return {
2301 'playbackContext': {
2302 'contentPlaybackContext': context
2303 },
2304 **cls._get_checkok_params()
2305 }
2306
2307 @staticmethod
2308 def _is_agegated(player_response):
2309 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
2310 return True
2311
2312 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
2313 AGE_GATE_REASONS = (
2314 'confirm your age', 'age-restricted', 'inappropriate', # reason
2315 'age_verification_required', 'age_check_required', # status
2316 )
2317 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
2318
2319 @staticmethod
2320 def _is_unplayable(player_response):
2321 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
2322
2323 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr):
2324
2325 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
2326 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
2327 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
2328 headers = self.generate_api_headers(
2329 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
2330
2331 yt_query = {'videoId': video_id}
2332 yt_query.update(self._generate_player_context(sts))
2333 return self._extract_response(
2334 item_id=video_id, ep='player', query=yt_query,
2335 ytcfg=player_ytcfg, headers=headers, fatal=True,
2336 default_client=client,
2337 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
2338 ) or None
2339
2340 def _get_requested_clients(self, url, smuggled_data):
2341 requested_clients = []
2342 allowed_clients = sorted(
2343 [client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'],
2344 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
2345 for client in self._configuration_arg('player_client'):
2346 if client in allowed_clients:
2347 requested_clients.append(client)
2348 elif client == 'all':
2349 requested_clients.extend(allowed_clients)
2350 else:
2351 self.report_warning(f'Skipping unsupported client {client}')
2352 if not requested_clients:
2353 requested_clients = ['android', 'web']
2354
2355 if smuggled_data.get('is_music_url') or self.is_music_url(url):
2356 requested_clients.extend(
2357 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
2358
2359 return orderedSet(requested_clients)
2360
2361 def _extract_player_ytcfg(self, client, video_id):
2362 url = {
2363 'web_music': 'https://music.youtube.com',
2364 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
2365 }.get(client)
2366 if not url:
2367 return {}
2368 webpage = self._download_webpage(url, video_id, fatal=False, note=f'Downloading {client} config')
2369 return self.extract_ytcfg(video_id, webpage) or {}
2370
2371 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
2372 initial_pr = None
2373 if webpage:
2374 initial_pr = self._extract_yt_initial_variable(
2375 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
2376 video_id, 'initial player response')
2377
2378 original_clients = clients
2379 clients = clients[::-1]
2380 prs = []
2381
2382 def append_client(client_name):
2383 if client_name in INNERTUBE_CLIENTS and client_name not in original_clients:
2384 clients.append(client_name)
2385
2386 # Android player_response does not have microFormats which are needed for
2387 # extraction of some data. So we return the initial_pr with formats
2388 # stripped out even if not requested by the user
2389 # See: https://github.com/yt-dlp/yt-dlp/issues/501
2390 if initial_pr:
2391 pr = dict(initial_pr)
2392 pr['streamingData'] = None
2393 prs.append(pr)
2394
2395 last_error = None
2396 tried_iframe_fallback = False
2397 player_url = None
2398 while clients:
2399 client = clients.pop()
2400 player_ytcfg = master_ytcfg if client == 'web' else {}
2401 if 'configs' not in self._configuration_arg('player_skip'):
2402 player_ytcfg = self._extract_player_ytcfg(client, video_id) or player_ytcfg
2403
2404 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
2405 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
2406 if 'js' in self._configuration_arg('player_skip'):
2407 require_js_player = False
2408 player_url = None
2409
2410 if not player_url and not tried_iframe_fallback and require_js_player:
2411 player_url = self._download_player_url(video_id)
2412 tried_iframe_fallback = True
2413
2414 try:
2415 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
2416 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr)
2417 except ExtractorError as e:
2418 if last_error:
2419 self.report_warning(last_error)
2420 last_error = e
2421 continue
2422
2423 if pr:
2424 prs.append(pr)
2425
2426 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
2427 if client.endswith('_agegate') and self._is_unplayable(pr) and self.is_authenticated:
2428 append_client(client.replace('_agegate', '_creator'))
2429 elif self._is_agegated(pr):
2430 append_client(f'{client}_agegate')
2431
2432 if last_error:
2433 if not len(prs):
2434 raise last_error
2435 self.report_warning(last_error)
2436 return prs, player_url
2437
2438 def _extract_formats(self, streaming_data, video_id, player_url, is_live):
2439 itags, stream_ids = {}, []
2440 itag_qualities, res_qualities = {}, {}
2441 q = qualities([
2442 # Normally tiny is the smallest video-only formats. But
2443 # audio-only formats with unknown quality may get tagged as tiny
2444 'tiny',
2445 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
2446 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
2447 ])
2448 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
2449
2450 for fmt in streaming_formats:
2451 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
2452 continue
2453
2454 itag = str_or_none(fmt.get('itag'))
2455 audio_track = fmt.get('audioTrack') or {}
2456 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
2457 if stream_id in stream_ids:
2458 continue
2459
2460 quality = fmt.get('quality')
2461 height = int_or_none(fmt.get('height'))
2462 if quality == 'tiny' or not quality:
2463 quality = fmt.get('audioQuality', '').lower() or quality
2464 # The 3gp format (17) in android client has a quality of "small",
2465 # but is actually worse than other formats
2466 if itag == '17':
2467 quality = 'tiny'
2468 if quality:
2469 if itag:
2470 itag_qualities[itag] = quality
2471 if height:
2472 res_qualities[height] = quality
2473 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
2474 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
2475 # number of fragment that would subsequently requested with (`&sq=N`)
2476 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
2477 continue
2478
2479 fmt_url = fmt.get('url')
2480 if not fmt_url:
2481 sc = compat_parse_qs(fmt.get('signatureCipher'))
2482 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
2483 encrypted_sig = try_get(sc, lambda x: x['s'][0])
2484 if not (sc and fmt_url and encrypted_sig):
2485 continue
2486 if not player_url:
2487 continue
2488 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
2489 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
2490 fmt_url += '&' + sp + '=' + signature
2491
2492 query = parse_qs(fmt_url)
2493 throttled = False
2494 if query.get('ratebypass') != ['yes'] and query.get('n'):
2495 try:
2496 fmt_url = update_url_query(fmt_url, {
2497 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)})
2498 except ExtractorError as e:
2499 self.report_warning(f'nsig extraction failed: You may experience throttling for some formats\n{e}', only_once=True)
2500 throttled = True
2501
2502 if itag:
2503 itags[itag] = 'https'
2504 stream_ids.append(stream_id)
2505
2506 tbr = float_or_none(
2507 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
2508 dct = {
2509 'asr': int_or_none(fmt.get('audioSampleRate')),
2510 'filesize': int_or_none(fmt.get('contentLength')),
2511 'format_id': itag,
2512 'format_note': join_nonempty(
2513 '%s%s' % (audio_track.get('displayName') or '',
2514 ' (default)' if audio_track.get('audioIsDefault') else ''),
2515 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
2516 throttled and 'THROTTLED', delim=', '),
2517 'source_preference': -10 if throttled else -1,
2518 'fps': int_or_none(fmt.get('fps')) or None,
2519 'height': height,
2520 'quality': q(quality),
2521 'tbr': tbr,
2522 'url': fmt_url,
2523 'width': int_or_none(fmt.get('width')),
2524 'language': audio_track.get('id', '').split('.')[0],
2525 'language_preference': 1 if audio_track.get('audioIsDefault') else -1,
2526 }
2527 mime_mobj = re.match(
2528 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
2529 if mime_mobj:
2530 dct['ext'] = mimetype2ext(mime_mobj.group(1))
2531 dct.update(parse_codecs(mime_mobj.group(2)))
2532 no_audio = dct.get('acodec') == 'none'
2533 no_video = dct.get('vcodec') == 'none'
2534 if no_audio:
2535 dct['vbr'] = tbr
2536 if no_video:
2537 dct['abr'] = tbr
2538 if no_audio or no_video:
2539 dct['downloader_options'] = {
2540 # Youtube throttles chunks >~10M
2541 'http_chunk_size': 10485760,
2542 }
2543 if dct.get('ext'):
2544 dct['container'] = dct['ext'] + '_dash'
2545 yield dct
2546
2547 skip_manifests = self._configuration_arg('skip')
2548 get_dash = (
2549 (not is_live or self._configuration_arg('include_live_dash'))
2550 and 'dash' not in skip_manifests and self.get_param('youtube_include_dash_manifest', True))
2551 get_hls = 'hls' not in skip_manifests and self.get_param('youtube_include_hls_manifest', True)
2552
2553 def process_manifest_format(f, proto, itag):
2554 if itag in itags:
2555 if itags[itag] == proto or f'{itag}-{proto}' in itags:
2556 return False
2557 itag = f'{itag}-{proto}'
2558 if itag:
2559 f['format_id'] = itag
2560 itags[itag] = proto
2561
2562 f['quality'] = next((
2563 q(qdict[val])
2564 for val, qdict in ((f.get('format_id'), itag_qualities), (f.get('height'), res_qualities))
2565 if val in qdict), -1)
2566 return True
2567
2568 for sd in streaming_data:
2569 hls_manifest_url = get_hls and sd.get('hlsManifestUrl')
2570 if hls_manifest_url:
2571 for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False):
2572 if process_manifest_format(f, 'hls', self._search_regex(
2573 r'/itag/(\d+)', f['url'], 'itag', default=None)):
2574 yield f
2575
2576 dash_manifest_url = get_dash and sd.get('dashManifestUrl')
2577 if dash_manifest_url:
2578 for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False):
2579 if process_manifest_format(f, 'dash', f['format_id']):
2580 f['filesize'] = int_or_none(self._search_regex(
2581 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
2582 yield f
2583
2584 def _real_extract(self, url):
2585 url, smuggled_data = unsmuggle_url(url, {})
2586 video_id = self._match_id(url)
2587
2588 base_url = self.http_scheme() + '//www.youtube.com/'
2589 webpage_url = base_url + 'watch?v=' + video_id
2590 webpage = None
2591 if 'webpage' not in self._configuration_arg('player_skip'):
2592 webpage = self._download_webpage(
2593 webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False)
2594
2595 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
2596
2597 player_responses, player_url = self._extract_player_responses(
2598 self._get_requested_clients(url, smuggled_data),
2599 video_id, webpage, master_ytcfg)
2600
2601 get_first = lambda obj, keys, **kwargs: traverse_obj(obj, (..., *variadic(keys)), **kwargs, get_all=False)
2602
2603 playability_statuses = traverse_obj(
2604 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
2605
2606 trailer_video_id = get_first(
2607 playability_statuses,
2608 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
2609 expected_type=str)
2610 if trailer_video_id:
2611 return self.url_result(
2612 trailer_video_id, self.ie_key(), trailer_video_id)
2613
2614 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
2615 if webpage else (lambda x: None))
2616
2617 video_details = traverse_obj(
2618 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
2619 microformats = traverse_obj(
2620 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
2621 expected_type=dict, default=[])
2622 video_title = (
2623 get_first(video_details, 'title')
2624 or self._get_text(microformats, (..., 'title'))
2625 or search_meta(['og:title', 'twitter:title', 'title']))
2626 video_description = get_first(video_details, 'shortDescription')
2627
2628 multifeed_metadata_list = get_first(
2629 player_responses,
2630 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
2631 expected_type=str)
2632 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
2633 if self.get_param('noplaylist'):
2634 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2635 else:
2636 entries = []
2637 feed_ids = []
2638 for feed in multifeed_metadata_list.split(','):
2639 # Unquote should take place before split on comma (,) since textual
2640 # fields may contain comma as well (see
2641 # https://github.com/ytdl-org/youtube-dl/issues/8536)
2642 feed_data = compat_parse_qs(
2643 compat_urllib_parse_unquote_plus(feed))
2644
2645 def feed_entry(name):
2646 return try_get(
2647 feed_data, lambda x: x[name][0], compat_str)
2648
2649 feed_id = feed_entry('id')
2650 if not feed_id:
2651 continue
2652 feed_title = feed_entry('title')
2653 title = video_title
2654 if feed_title:
2655 title += ' (%s)' % feed_title
2656 entries.append({
2657 '_type': 'url_transparent',
2658 'ie_key': 'Youtube',
2659 'url': smuggle_url(
2660 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
2661 {'force_singlefeed': True}),
2662 'title': title,
2663 })
2664 feed_ids.append(feed_id)
2665 self.to_screen(
2666 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
2667 % (', '.join(feed_ids), video_id))
2668 return self.playlist_result(
2669 entries, video_id, video_title, video_description)
2670
2671 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
2672 is_live = get_first(video_details, 'isLive')
2673 if is_live is None:
2674 is_live = get_first(live_broadcast_details, 'isLiveNow')
2675
2676 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
2677 formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live))
2678
2679 if not formats:
2680 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
2681 self.report_drm(video_id)
2682 pemr = get_first(
2683 playability_statuses,
2684 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
2685 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
2686 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
2687 if subreason:
2688 if subreason == 'The uploader has not made this video available in your country.':
2689 countries = get_first(microformats, 'availableCountries')
2690 if not countries:
2691 regions_allowed = search_meta('regionsAllowed')
2692 countries = regions_allowed.split(',') if regions_allowed else None
2693 self.raise_geo_restricted(subreason, countries, metadata_available=True)
2694 reason += f'. {subreason}'
2695 if reason:
2696 self.raise_no_formats(reason, expected=True)
2697
2698 # Source is given priority since formats that throttle are given lower source_preference
2699 # When throttling issue is fully fixed, remove this
2700 self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang', 'proto'))
2701
2702 keywords = get_first(video_details, 'keywords', expected_type=list) or []
2703 if not keywords and webpage:
2704 keywords = [
2705 unescapeHTML(m.group('content'))
2706 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
2707 for keyword in keywords:
2708 if keyword.startswith('yt:stretch='):
2709 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
2710 if mobj:
2711 # NB: float is intentional for forcing float division
2712 w, h = (float(v) for v in mobj.groups())
2713 if w > 0 and h > 0:
2714 ratio = w / h
2715 for f in formats:
2716 if f.get('vcodec') != 'none':
2717 f['stretched_ratio'] = ratio
2718 break
2719
2720 thumbnails = []
2721 thumbnail_dicts = traverse_obj(
2722 (video_details, microformats), (..., ..., 'thumbnail', 'thumbnails', ...),
2723 expected_type=dict, default=[])
2724 for thumbnail in thumbnail_dicts:
2725 thumbnail_url = thumbnail.get('url')
2726 if not thumbnail_url:
2727 continue
2728 # Sometimes youtube gives a wrong thumbnail URL. See:
2729 # https://github.com/yt-dlp/yt-dlp/issues/233
2730 # https://github.com/ytdl-org/youtube-dl/issues/28023
2731 if 'maxresdefault' in thumbnail_url:
2732 thumbnail_url = thumbnail_url.split('?')[0]
2733 thumbnails.append({
2734 'url': thumbnail_url,
2735 'height': int_or_none(thumbnail.get('height')),
2736 'width': int_or_none(thumbnail.get('width')),
2737 })
2738 thumbnail_url = search_meta(['og:image', 'twitter:image'])
2739 if thumbnail_url:
2740 thumbnails.append({
2741 'url': thumbnail_url,
2742 })
2743 original_thumbnails = thumbnails.copy()
2744
2745 # The best resolution thumbnails sometimes does not appear in the webpage
2746 # See: https://github.com/ytdl-org/youtube-dl/issues/29049, https://github.com/yt-dlp/yt-dlp/issues/340
2747 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
2748 thumbnail_names = [
2749 'maxresdefault', 'hq720', 'sddefault', 'sd1', 'sd2', 'sd3',
2750 'hqdefault', 'hq1', 'hq2', 'hq3', '0',
2751 'mqdefault', 'mq1', 'mq2', 'mq3',
2752 'default', '1', '2', '3'
2753 ]
2754 n_thumbnail_names = len(thumbnail_names)
2755 thumbnails.extend({
2756 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
2757 video_id=video_id, name=name, ext=ext,
2758 webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''),
2759 } for name in thumbnail_names for ext in ('webp', 'jpg'))
2760 for thumb in thumbnails:
2761 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
2762 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
2763 self._remove_duplicate_formats(thumbnails)
2764 self._downloader._sort_thumbnails(original_thumbnails)
2765
2766 category = get_first(microformats, 'category') or search_meta('genre')
2767 channel_id = str_or_none(
2768 get_first(video_details, 'channelId')
2769 or get_first(microformats, 'externalChannelId')
2770 or search_meta('channelId'))
2771 duration = int_or_none(
2772 get_first(video_details, 'lengthSeconds')
2773 or get_first(microformats, 'lengthSeconds')
2774 or parse_duration(search_meta('duration'))) or None
2775 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
2776
2777 live_content = get_first(video_details, 'isLiveContent')
2778 is_upcoming = get_first(video_details, 'isUpcoming')
2779 if is_live is None:
2780 if is_upcoming or live_content is False:
2781 is_live = False
2782 if is_upcoming is None and (live_content or is_live):
2783 is_upcoming = False
2784 live_starttime = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
2785 live_endtime = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
2786 if not duration and live_endtime and live_starttime:
2787 duration = live_endtime - live_starttime
2788
2789 info = {
2790 'id': video_id,
2791 'title': self._live_title(video_title) if is_live else video_title,
2792 'formats': formats,
2793 'thumbnails': thumbnails,
2794 # The best thumbnail that we are sure exists. Prevents unnecessary
2795 # URL checking if user don't care about getting the best possible thumbnail
2796 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
2797 'description': video_description,
2798 'upload_date': unified_strdate(
2799 get_first(microformats, 'uploadDate')
2800 or search_meta('uploadDate')),
2801 'uploader': get_first(video_details, 'author'),
2802 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
2803 'uploader_url': owner_profile_url,
2804 'channel_id': channel_id,
2805 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
2806 'duration': duration,
2807 'view_count': int_or_none(
2808 get_first((video_details, microformats), (..., 'viewCount'))
2809 or search_meta('interactionCount')),
2810 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
2811 'age_limit': 18 if (
2812 get_first(microformats, 'isFamilySafe') is False
2813 or search_meta('isFamilyFriendly') == 'false'
2814 or search_meta('og:restrictions:age') == '18+') else 0,
2815 'webpage_url': webpage_url,
2816 'categories': [category] if category else None,
2817 'tags': keywords,
2818 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
2819 'is_live': is_live,
2820 'was_live': (False if is_live or is_upcoming or live_content is False
2821 else None if is_live is None or is_upcoming is None
2822 else live_content),
2823 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
2824 'release_timestamp': live_starttime,
2825 }
2826
2827 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
2828 if pctr:
2829 def get_lang_code(track):
2830 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
2831 or track.get('languageCode'))
2832
2833 # Converted into dicts to remove duplicates
2834 captions = {
2835 get_lang_code(sub): sub
2836 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
2837 translation_languages = {
2838 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
2839 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
2840
2841 def process_language(container, base_url, lang_code, sub_name, query):
2842 lang_subs = container.setdefault(lang_code, [])
2843 for fmt in self._SUBTITLE_FORMATS:
2844 query.update({
2845 'fmt': fmt,
2846 })
2847 lang_subs.append({
2848 'ext': fmt,
2849 'url': update_url_query(base_url, query),
2850 'name': sub_name,
2851 })
2852
2853 subtitles, automatic_captions = {}, {}
2854 for lang_code, caption_track in captions.items():
2855 base_url = caption_track.get('baseUrl')
2856 if not base_url:
2857 continue
2858 lang_name = self._get_text(caption_track, 'name', max_runs=1)
2859 if caption_track.get('kind') != 'asr':
2860 if not lang_code:
2861 continue
2862 process_language(
2863 subtitles, base_url, lang_code, lang_name, {})
2864 if not caption_track.get('isTranslatable'):
2865 continue
2866 for trans_code, trans_name in translation_languages.items():
2867 if not trans_code:
2868 continue
2869 if caption_track.get('kind') != 'asr':
2870 trans_code += f'-{lang_code}'
2871 trans_name += format_field(lang_name, template=' from %s')
2872 process_language(
2873 automatic_captions, base_url, trans_code, trans_name, {'tlang': trans_code})
2874 info['automatic_captions'] = automatic_captions
2875 info['subtitles'] = subtitles
2876
2877 parsed_url = compat_urllib_parse_urlparse(url)
2878 for component in [parsed_url.fragment, parsed_url.query]:
2879 query = compat_parse_qs(component)
2880 for k, v in query.items():
2881 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
2882 d_k += '_time'
2883 if d_k not in info and k in s_ks:
2884 info[d_k] = parse_duration(query[k][0])
2885
2886 # Youtube Music Auto-generated description
2887 if video_description:
2888 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)
2889 if mobj:
2890 release_year = mobj.group('release_year')
2891 release_date = mobj.group('release_date')
2892 if release_date:
2893 release_date = release_date.replace('-', '')
2894 if not release_year:
2895 release_year = release_date[:4]
2896 info.update({
2897 'album': mobj.group('album'.strip()),
2898 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
2899 'track': mobj.group('track').strip(),
2900 'release_date': release_date,
2901 'release_year': int_or_none(release_year),
2902 })
2903
2904 initial_data = None
2905 if webpage:
2906 initial_data = self._extract_yt_initial_variable(
2907 webpage, self._YT_INITIAL_DATA_RE, video_id,
2908 'yt initial data')
2909 if not initial_data:
2910 query = {'videoId': video_id}
2911 query.update(self._get_checkok_params())
2912 initial_data = self._extract_response(
2913 item_id=video_id, ep='next', fatal=False,
2914 ytcfg=master_ytcfg, query=query,
2915 headers=self.generate_api_headers(ytcfg=master_ytcfg),
2916 note='Downloading initial data API JSON')
2917
2918 try:
2919 # This will error if there is no livechat
2920 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2921 info.setdefault('subtitles', {})['live_chat'] = [{
2922 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
2923 'video_id': video_id,
2924 'ext': 'json',
2925 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay',
2926 }]
2927 except (KeyError, IndexError, TypeError):
2928 pass
2929
2930 if initial_data:
2931 info['chapters'] = (
2932 self._extract_chapters_from_json(initial_data, duration)
2933 or self._extract_chapters_from_engagement_panel(initial_data, duration)
2934 or None)
2935
2936 contents = try_get(
2937 initial_data,
2938 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
2939 list) or []
2940 for content in contents:
2941 vpir = content.get('videoPrimaryInfoRenderer')
2942 if vpir:
2943 stl = vpir.get('superTitleLink')
2944 if stl:
2945 stl = self._get_text(stl)
2946 if try_get(
2947 vpir,
2948 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
2949 info['location'] = stl
2950 else:
2951 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
2952 if mobj:
2953 info.update({
2954 'series': mobj.group(1),
2955 'season_number': int(mobj.group(2)),
2956 'episode_number': int(mobj.group(3)),
2957 })
2958 for tlb in (try_get(
2959 vpir,
2960 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
2961 list) or []):
2962 tbr = tlb.get('toggleButtonRenderer') or {}
2963 for getter, regex in [(
2964 lambda x: x['defaultText']['accessibility']['accessibilityData'],
2965 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
2966 lambda x: x['accessibility'],
2967 lambda x: x['accessibilityData']['accessibilityData'],
2968 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
2969 label = (try_get(tbr, getter, dict) or {}).get('label')
2970 if label:
2971 mobj = re.match(regex, label)
2972 if mobj:
2973 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
2974 break
2975 sbr_tooltip = try_get(
2976 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
2977 if sbr_tooltip:
2978 like_count, dislike_count = sbr_tooltip.split(' / ')
2979 info.update({
2980 'like_count': str_to_int(like_count),
2981 'dislike_count': str_to_int(dislike_count),
2982 })
2983 vsir = content.get('videoSecondaryInfoRenderer')
2984 if vsir:
2985 info['channel'] = self._get_text(vsir, ('owner', 'videoOwnerRenderer', 'title'))
2986 rows = try_get(
2987 vsir,
2988 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
2989 list) or []
2990 multiple_songs = False
2991 for row in rows:
2992 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2993 multiple_songs = True
2994 break
2995 for row in rows:
2996 mrr = row.get('metadataRowRenderer') or {}
2997 mrr_title = mrr.get('title')
2998 if not mrr_title:
2999 continue
3000 mrr_title = self._get_text(mrr, 'title')
3001 mrr_contents_text = self._get_text(mrr, ('contents', 0))
3002 if mrr_title == 'License':
3003 info['license'] = mrr_contents_text
3004 elif not multiple_songs:
3005 if mrr_title == 'Album':
3006 info['album'] = mrr_contents_text
3007 elif mrr_title == 'Artist':
3008 info['artist'] = mrr_contents_text
3009 elif mrr_title == 'Song':
3010 info['track'] = mrr_contents_text
3011
3012 fallbacks = {
3013 'channel': 'uploader',
3014 'channel_id': 'uploader_id',
3015 'channel_url': 'uploader_url',
3016 }
3017 for to, frm in fallbacks.items():
3018 if not info.get(to):
3019 info[to] = info.get(frm)
3020
3021 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
3022 v = info.get(s_k)
3023 if v:
3024 info[d_k] = v
3025
3026 is_private = get_first(video_details, 'isPrivate', expected_type=bool)
3027 is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool)
3028 is_membersonly = None
3029 is_premium = None
3030 if initial_data and is_private is not None:
3031 is_membersonly = False
3032 is_premium = False
3033 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
3034 badge_labels = set()
3035 for content in contents:
3036 if not isinstance(content, dict):
3037 continue
3038 badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
3039 for badge_label in badge_labels:
3040 if badge_label.lower() == 'members only':
3041 is_membersonly = True
3042 elif badge_label.lower() == 'premium':
3043 is_premium = True
3044 elif badge_label.lower() == 'unlisted':
3045 is_unlisted = True
3046
3047 info['availability'] = self._availability(
3048 is_private=is_private,
3049 needs_premium=is_premium,
3050 needs_subscription=is_membersonly,
3051 needs_auth=info['age_limit'] >= 18,
3052 is_unlisted=None if is_private is None else is_unlisted)
3053
3054 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
3055
3056 self.mark_watched(video_id, player_responses)
3057
3058 return info
3059
3060
3061 class YoutubeTabIE(YoutubeBaseInfoExtractor):
3062 IE_DESC = 'YouTube Tabs'
3063 _VALID_URL = r'''(?x)
3064 https?://
3065 (?:\w+\.)?
3066 (?:
3067 youtube(?:kids)?\.com|
3068 %(invidious)s
3069 )/
3070 (?:
3071 (?P<channel_type>channel|c|user|browse)/|
3072 (?P<not_channel>
3073 feed/|hashtag/|
3074 (?:playlist|watch)\?.*?\blist=
3075 )|
3076 (?!(?:%(reserved_names)s)\b) # Direct URLs
3077 )
3078 (?P<id>[^/?\#&]+)
3079 ''' % {
3080 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
3081 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
3082 }
3083 IE_NAME = 'youtube:tab'
3084
3085 _TESTS = [{
3086 'note': 'playlists, multipage',
3087 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
3088 'playlist_mincount': 94,
3089 'info_dict': {
3090 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
3091 'title': 'Игорь Клейнер - Playlists',
3092 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
3093 'uploader': 'Игорь Клейнер',
3094 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
3095 },
3096 }, {
3097 'note': 'playlists, multipage, different order',
3098 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3099 'playlist_mincount': 94,
3100 'info_dict': {
3101 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
3102 'title': 'Игорь Клейнер - Playlists',
3103 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
3104 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
3105 'uploader': 'Игорь Клейнер',
3106 },
3107 }, {
3108 'note': 'playlists, series',
3109 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
3110 'playlist_mincount': 5,
3111 'info_dict': {
3112 'id': 'UCYO_jab_esuFRV4b17AJtAw',
3113 'title': '3Blue1Brown - Playlists',
3114 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
3115 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
3116 'uploader': '3Blue1Brown',
3117 },
3118 }, {
3119 'note': 'playlists, singlepage',
3120 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
3121 'playlist_mincount': 4,
3122 'info_dict': {
3123 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
3124 'title': 'ThirstForScience - Playlists',
3125 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
3126 'uploader': 'ThirstForScience',
3127 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
3128 }
3129 }, {
3130 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
3131 'only_matching': True,
3132 }, {
3133 'note': 'basic, single video playlist',
3134 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
3135 'info_dict': {
3136 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
3137 'uploader': 'Sergey M.',
3138 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
3139 'title': 'youtube-dl public playlist',
3140 },
3141 'playlist_count': 1,
3142 }, {
3143 'note': 'empty playlist',
3144 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
3145 'info_dict': {
3146 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
3147 'uploader': 'Sergey M.',
3148 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
3149 'title': 'youtube-dl empty playlist',
3150 },
3151 'playlist_count': 0,
3152 }, {
3153 'note': 'Home tab',
3154 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
3155 'info_dict': {
3156 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3157 'title': 'lex will - Home',
3158 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
3159 'uploader': 'lex will',
3160 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3161 },
3162 'playlist_mincount': 2,
3163 }, {
3164 'note': 'Videos tab',
3165 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
3166 'info_dict': {
3167 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3168 'title': 'lex will - Videos',
3169 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
3170 'uploader': 'lex will',
3171 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3172 },
3173 'playlist_mincount': 975,
3174 }, {
3175 'note': 'Videos tab, sorted by popular',
3176 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
3177 'info_dict': {
3178 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3179 'title': 'lex will - Videos',
3180 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
3181 'uploader': 'lex will',
3182 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3183 },
3184 'playlist_mincount': 199,
3185 }, {
3186 'note': 'Playlists tab',
3187 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
3188 'info_dict': {
3189 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3190 'title': 'lex will - Playlists',
3191 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
3192 'uploader': 'lex will',
3193 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3194 },
3195 'playlist_mincount': 17,
3196 }, {
3197 'note': 'Community tab',
3198 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
3199 'info_dict': {
3200 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3201 'title': 'lex will - Community',
3202 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
3203 'uploader': 'lex will',
3204 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3205 },
3206 'playlist_mincount': 18,
3207 }, {
3208 'note': 'Channels tab',
3209 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
3210 'info_dict': {
3211 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3212 'title': 'lex will - Channels',
3213 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
3214 'uploader': 'lex will',
3215 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3216 },
3217 'playlist_mincount': 12,
3218 }, {
3219 'note': 'Search tab',
3220 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
3221 'playlist_mincount': 40,
3222 'info_dict': {
3223 'id': 'UCYO_jab_esuFRV4b17AJtAw',
3224 'title': '3Blue1Brown - Search - linear algebra',
3225 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
3226 'uploader': '3Blue1Brown',
3227 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
3228 },
3229 }, {
3230 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
3231 'only_matching': True,
3232 }, {
3233 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
3234 'only_matching': True,
3235 }, {
3236 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
3237 'only_matching': True,
3238 }, {
3239 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
3240 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
3241 'info_dict': {
3242 'title': '29C3: Not my department',
3243 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
3244 'uploader': 'Christiaan008',
3245 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
3246 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
3247 },
3248 'playlist_count': 96,
3249 }, {
3250 'note': 'Large playlist',
3251 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
3252 'info_dict': {
3253 'title': 'Uploads from Cauchemar',
3254 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
3255 'uploader': 'Cauchemar',
3256 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
3257 },
3258 'playlist_mincount': 1123,
3259 }, {
3260 'note': 'even larger playlist, 8832 videos',
3261 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
3262 'only_matching': True,
3263 }, {
3264 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
3265 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
3266 'info_dict': {
3267 'title': 'Uploads from Interstellar Movie',
3268 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
3269 'uploader': 'Interstellar Movie',
3270 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
3271 },
3272 'playlist_mincount': 21,
3273 }, {
3274 'note': 'Playlist with "show unavailable videos" button',
3275 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
3276 'info_dict': {
3277 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
3278 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
3279 'uploader': 'Phim Siêu Nhân Nhật Bản',
3280 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
3281 },
3282 'playlist_mincount': 200,
3283 }, {
3284 'note': 'Playlist with unavailable videos in page 7',
3285 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
3286 'info_dict': {
3287 'title': 'Uploads from BlankTV',
3288 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
3289 'uploader': 'BlankTV',
3290 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
3291 },
3292 'playlist_mincount': 1000,
3293 }, {
3294 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
3295 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
3296 'info_dict': {
3297 'title': 'Data Analysis with Dr Mike Pound',
3298 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
3299 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
3300 'uploader': 'Computerphile',
3301 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
3302 },
3303 'playlist_mincount': 11,
3304 }, {
3305 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
3306 'only_matching': True,
3307 }, {
3308 'note': 'Playlist URL that does not actually serve a playlist',
3309 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
3310 'info_dict': {
3311 'id': 'FqZTN594JQw',
3312 'ext': 'webm',
3313 'title': "Smiley's People 01 detective, Adventure Series, Action",
3314 'uploader': 'STREEM',
3315 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
3316 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
3317 'upload_date': '20150526',
3318 'license': 'Standard YouTube License',
3319 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
3320 'categories': ['People & Blogs'],
3321 'tags': list,
3322 'view_count': int,
3323 'like_count': int,
3324 'dislike_count': int,
3325 },
3326 'params': {
3327 'skip_download': True,
3328 },
3329 'skip': 'This video is not available.',
3330 'add_ie': [YoutubeIE.ie_key()],
3331 }, {
3332 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
3333 'only_matching': True,
3334 }, {
3335 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
3336 'only_matching': True,
3337 }, {
3338 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
3339 'info_dict': {
3340 'id': '3yImotZU3tw', # This will keep changing
3341 'ext': 'mp4',
3342 'title': compat_str,
3343 'uploader': 'Sky News',
3344 'uploader_id': 'skynews',
3345 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
3346 'upload_date': r're:\d{8}',
3347 'description': compat_str,
3348 'categories': ['News & Politics'],
3349 'tags': list,
3350 'like_count': int,
3351 'dislike_count': int,
3352 },
3353 'params': {
3354 'skip_download': True,
3355 },
3356 'expected_warnings': ['Downloading just video ', 'Ignoring subtitle tracks found in '],
3357 }, {
3358 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
3359 'info_dict': {
3360 'id': 'a48o2S1cPoo',
3361 'ext': 'mp4',
3362 'title': 'The Young Turks - Live Main Show',
3363 'uploader': 'The Young Turks',
3364 'uploader_id': 'TheYoungTurks',
3365 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
3366 'upload_date': '20150715',
3367 'license': 'Standard YouTube License',
3368 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
3369 'categories': ['News & Politics'],
3370 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
3371 'like_count': int,
3372 'dislike_count': int,
3373 },
3374 'params': {
3375 'skip_download': True,
3376 },
3377 'only_matching': True,
3378 }, {
3379 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
3380 'only_matching': True,
3381 }, {
3382 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
3383 'only_matching': True,
3384 }, {
3385 'note': 'A channel that is not live. Should raise error',
3386 'url': 'https://www.youtube.com/user/numberphile/live',
3387 'only_matching': True,
3388 }, {
3389 'url': 'https://www.youtube.com/feed/trending',
3390 'only_matching': True,
3391 }, {
3392 'url': 'https://www.youtube.com/feed/library',
3393 'only_matching': True,
3394 }, {
3395 'url': 'https://www.youtube.com/feed/history',
3396 'only_matching': True,
3397 }, {
3398 'url': 'https://www.youtube.com/feed/subscriptions',
3399 'only_matching': True,
3400 }, {
3401 'url': 'https://www.youtube.com/feed/watch_later',
3402 'only_matching': True,
3403 }, {
3404 'note': 'Recommended - redirects to home page.',
3405 'url': 'https://www.youtube.com/feed/recommended',
3406 'only_matching': True,
3407 }, {
3408 'note': 'inline playlist with not always working continuations',
3409 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
3410 'only_matching': True,
3411 }, {
3412 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8',
3413 'only_matching': True,
3414 }, {
3415 'url': 'https://www.youtube.com/course',
3416 'only_matching': True,
3417 }, {
3418 'url': 'https://www.youtube.com/zsecurity',
3419 'only_matching': True,
3420 }, {
3421 'url': 'http://www.youtube.com/NASAgovVideo/videos',
3422 'only_matching': True,
3423 }, {
3424 'url': 'https://www.youtube.com/TheYoungTurks/live',
3425 'only_matching': True,
3426 }, {
3427 'url': 'https://www.youtube.com/hashtag/cctv9',
3428 'info_dict': {
3429 'id': 'cctv9',
3430 'title': '#cctv9',
3431 },
3432 'playlist_mincount': 350,
3433 }, {
3434 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
3435 'only_matching': True,
3436 }, {
3437 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
3438 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
3439 'only_matching': True
3440 }, {
3441 'note': '/browse/ should redirect to /channel/',
3442 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
3443 'only_matching': True
3444 }, {
3445 'note': 'VLPL, should redirect to playlist?list=PL...',
3446 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
3447 'info_dict': {
3448 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
3449 'uploader': 'NoCopyrightSounds',
3450 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
3451 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
3452 'title': 'NCS Releases',
3453 },
3454 'playlist_mincount': 166,
3455 }, {
3456 'note': 'Topic, should redirect to playlist?list=UU...',
3457 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
3458 'info_dict': {
3459 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
3460 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
3461 'title': 'Uploads from Royalty Free Music - Topic',
3462 'uploader': 'Royalty Free Music - Topic',
3463 },
3464 'expected_warnings': [
3465 'A channel/user page was given',
3466 'The URL does not have a videos tab',
3467 ],
3468 'playlist_mincount': 101,
3469 }, {
3470 'note': 'Topic without a UU playlist',
3471 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
3472 'info_dict': {
3473 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
3474 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
3475 },
3476 'expected_warnings': [
3477 'A channel/user page was given',
3478 'The URL does not have a videos tab',
3479 'Falling back to channel URL',
3480 ],
3481 'playlist_mincount': 9,
3482 }, {
3483 'note': 'Youtube music Album',
3484 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
3485 'info_dict': {
3486 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
3487 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
3488 },
3489 'playlist_count': 50,
3490 }, {
3491 'note': 'unlisted single video playlist',
3492 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
3493 'info_dict': {
3494 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
3495 'uploader': 'colethedj',
3496 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
3497 'title': 'yt-dlp unlisted playlist test',
3498 'availability': 'unlisted'
3499 },
3500 'playlist_count': 1,
3501 }, {
3502 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
3503 'url': 'https://www.youtube.com/feed/recommended',
3504 'info_dict': {
3505 'id': 'recommended',
3506 'title': 'recommended',
3507 },
3508 'playlist_mincount': 50,
3509 'params': {
3510 'skip_download': True,
3511 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
3512 },
3513 }, {
3514 'note': 'API Fallback: /videos tab, sorted by oldest first',
3515 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
3516 'info_dict': {
3517 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
3518 'title': 'Cody\'sLab - Videos',
3519 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
3520 'uploader': 'Cody\'sLab',
3521 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
3522 },
3523 'playlist_mincount': 650,
3524 'params': {
3525 'skip_download': True,
3526 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
3527 },
3528 }, {
3529 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
3530 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
3531 'info_dict': {
3532 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
3533 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
3534 'title': 'Uploads from Royalty Free Music - Topic',
3535 'uploader': 'Royalty Free Music - Topic',
3536 },
3537 'expected_warnings': [
3538 'A channel/user page was given',
3539 'The URL does not have a videos tab',
3540 ],
3541 'playlist_mincount': 101,
3542 'params': {
3543 'skip_download': True,
3544 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
3545 },
3546 }]
3547
3548 @classmethod
3549 def suitable(cls, url):
3550 return False if YoutubeIE.suitable(url) else super(
3551 YoutubeTabIE, cls).suitable(url)
3552
3553 def _extract_channel_id(self, webpage):
3554 channel_id = self._html_search_meta(
3555 'channelId', webpage, 'channel id', default=None)
3556 if channel_id:
3557 return channel_id
3558 channel_url = self._html_search_meta(
3559 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
3560 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
3561 'twitter:app:url:googleplay'), webpage, 'channel url')
3562 return self._search_regex(
3563 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
3564 channel_url, 'channel id')
3565
3566 @staticmethod
3567 def _extract_basic_item_renderer(item):
3568 # Modified from _extract_grid_item_renderer
3569 known_basic_renderers = (
3570 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer'
3571 )
3572 for key, renderer in item.items():
3573 if not isinstance(renderer, dict):
3574 continue
3575 elif key in known_basic_renderers:
3576 return renderer
3577 elif key.startswith('grid') and key.endswith('Renderer'):
3578 return renderer
3579
3580 def _grid_entries(self, grid_renderer):
3581 for item in grid_renderer['items']:
3582 if not isinstance(item, dict):
3583 continue
3584 renderer = self._extract_basic_item_renderer(item)
3585 if not isinstance(renderer, dict):
3586 continue
3587 title = self._get_text(renderer, 'title')
3588
3589 # playlist
3590 playlist_id = renderer.get('playlistId')
3591 if playlist_id:
3592 yield self.url_result(
3593 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3594 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3595 video_title=title)
3596 continue
3597 # video
3598 video_id = renderer.get('videoId')
3599 if video_id:
3600 yield self._extract_video(renderer)
3601 continue
3602 # channel
3603 channel_id = renderer.get('channelId')
3604 if channel_id:
3605 yield self.url_result(
3606 'https://www.youtube.com/channel/%s' % channel_id,
3607 ie=YoutubeTabIE.ie_key(), video_title=title)
3608 continue
3609 # generic endpoint URL support
3610 ep_url = urljoin('https://www.youtube.com/', try_get(
3611 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
3612 compat_str))
3613 if ep_url:
3614 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
3615 if ie.suitable(ep_url):
3616 yield self.url_result(
3617 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
3618 break
3619
3620 def _shelf_entries_from_content(self, shelf_renderer):
3621 content = shelf_renderer.get('content')
3622 if not isinstance(content, dict):
3623 return
3624 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
3625 if renderer:
3626 # TODO: add support for nested playlists so each shelf is processed
3627 # as separate playlist
3628 # TODO: this includes only first N items
3629 for entry in self._grid_entries(renderer):
3630 yield entry
3631 renderer = content.get('horizontalListRenderer')
3632 if renderer:
3633 # TODO
3634 pass
3635
3636 def _shelf_entries(self, shelf_renderer, skip_channels=False):
3637 ep = try_get(
3638 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3639 compat_str)
3640 shelf_url = urljoin('https://www.youtube.com', ep)
3641 if shelf_url:
3642 # Skipping links to another channels, note that checking for
3643 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
3644 # will not work
3645 if skip_channels and '/channels?' in shelf_url:
3646 return
3647 title = self._get_text(shelf_renderer, 'title')
3648 yield self.url_result(shelf_url, video_title=title)
3649 # Shelf may not contain shelf URL, fallback to extraction from content
3650 for entry in self._shelf_entries_from_content(shelf_renderer):
3651 yield entry
3652
3653 def _playlist_entries(self, video_list_renderer):
3654 for content in video_list_renderer['contents']:
3655 if not isinstance(content, dict):
3656 continue
3657 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
3658 if not isinstance(renderer, dict):
3659 continue
3660 video_id = renderer.get('videoId')
3661 if not video_id:
3662 continue
3663 yield self._extract_video(renderer)
3664
3665 def _rich_entries(self, rich_grid_renderer):
3666 renderer = try_get(
3667 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
3668 video_id = renderer.get('videoId')
3669 if not video_id:
3670 return
3671 yield self._extract_video(renderer)
3672
3673 def _video_entry(self, video_renderer):
3674 video_id = video_renderer.get('videoId')
3675 if video_id:
3676 return self._extract_video(video_renderer)
3677
3678 def _post_thread_entries(self, post_thread_renderer):
3679 post_renderer = try_get(
3680 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
3681 if not post_renderer:
3682 return
3683 # video attachment
3684 video_renderer = try_get(
3685 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
3686 video_id = video_renderer.get('videoId')
3687 if video_id:
3688 entry = self._extract_video(video_renderer)
3689 if entry:
3690 yield entry
3691 # playlist attachment
3692 playlist_id = try_get(
3693 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
3694 if playlist_id:
3695 yield self.url_result(
3696 'https://www.youtube.com/playlist?list=%s' % playlist_id,
3697 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3698 # inline video links
3699 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
3700 for run in runs:
3701 if not isinstance(run, dict):
3702 continue
3703 ep_url = try_get(
3704 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
3705 if not ep_url:
3706 continue
3707 if not YoutubeIE.suitable(ep_url):
3708 continue
3709 ep_video_id = YoutubeIE._match_id(ep_url)
3710 if video_id == ep_video_id:
3711 continue
3712 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
3713
3714 def _post_thread_continuation_entries(self, post_thread_continuation):
3715 contents = post_thread_continuation.get('contents')
3716 if not isinstance(contents, list):
3717 return
3718 for content in contents:
3719 renderer = content.get('backstagePostThreadRenderer')
3720 if not isinstance(renderer, dict):
3721 continue
3722 for entry in self._post_thread_entries(renderer):
3723 yield entry
3724
3725 r''' # unused
3726 def _rich_grid_entries(self, contents):
3727 for content in contents:
3728 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
3729 if video_renderer:
3730 entry = self._video_entry(video_renderer)
3731 if entry:
3732 yield entry
3733 '''
3734 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
3735
3736 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
3737 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
3738 for content in contents:
3739 if not isinstance(content, dict):
3740 continue
3741 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
3742 if not is_renderer:
3743 renderer = content.get('richItemRenderer')
3744 if renderer:
3745 for entry in self._rich_entries(renderer):
3746 yield entry
3747 continuation_list[0] = self._extract_continuation(parent_renderer)
3748 continue
3749 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
3750 for isr_content in isr_contents:
3751 if not isinstance(isr_content, dict):
3752 continue
3753
3754 known_renderers = {
3755 'playlistVideoListRenderer': self._playlist_entries,
3756 'gridRenderer': self._grid_entries,
3757 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'),
3758 'backstagePostThreadRenderer': self._post_thread_entries,
3759 'videoRenderer': lambda x: [self._video_entry(x)],
3760 }
3761 for key, renderer in isr_content.items():
3762 if key not in known_renderers:
3763 continue
3764 for entry in known_renderers[key](renderer):
3765 if entry:
3766 yield entry
3767 continuation_list[0] = self._extract_continuation(renderer)
3768 break
3769
3770 if not continuation_list[0]:
3771 continuation_list[0] = self._extract_continuation(is_renderer)
3772
3773 if not continuation_list[0]:
3774 continuation_list[0] = self._extract_continuation(parent_renderer)
3775
3776 continuation_list = [None] # Python 2 does not support nonlocal
3777 tab_content = try_get(tab, lambda x: x['content'], dict)
3778 if not tab_content:
3779 return
3780 parent_renderer = (
3781 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
3782 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
3783 for entry in extract_entries(parent_renderer):
3784 yield entry
3785 continuation = continuation_list[0]
3786
3787 for page_num in itertools.count(1):
3788 if not continuation:
3789 break
3790 headers = self.generate_api_headers(
3791 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
3792 response = self._extract_response(
3793 item_id='%s page %s' % (item_id, page_num),
3794 query=continuation, headers=headers, ytcfg=ytcfg,
3795 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3796
3797 if not response:
3798 break
3799 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
3800 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
3801 visitor_data = self._extract_visitor_data(response) or visitor_data
3802
3803 known_continuation_renderers = {
3804 'playlistVideoListContinuation': self._playlist_entries,
3805 'gridContinuation': self._grid_entries,
3806 'itemSectionContinuation': self._post_thread_continuation_entries,
3807 'sectionListContinuation': extract_entries, # for feeds
3808 }
3809 continuation_contents = try_get(
3810 response, lambda x: x['continuationContents'], dict) or {}
3811 continuation_renderer = None
3812 for key, value in continuation_contents.items():
3813 if key not in known_continuation_renderers:
3814 continue
3815 continuation_renderer = value
3816 continuation_list = [None]
3817 for entry in known_continuation_renderers[key](continuation_renderer):
3818 yield entry
3819 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
3820 break
3821 if continuation_renderer:
3822 continue
3823
3824 known_renderers = {
3825 'gridPlaylistRenderer': (self._grid_entries, 'items'),
3826 'gridVideoRenderer': (self._grid_entries, 'items'),
3827 'gridChannelRenderer': (self._grid_entries, 'items'),
3828 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
3829 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
3830 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
3831 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
3832 }
3833 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3834 continuation_items = try_get(
3835 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
3836 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
3837 video_items_renderer = None
3838 for key, value in continuation_item.items():
3839 if key not in known_renderers:
3840 continue
3841 video_items_renderer = {known_renderers[key][1]: continuation_items}
3842 continuation_list = [None]
3843 for entry in known_renderers[key][0](video_items_renderer):
3844 yield entry
3845 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3846 break
3847 if video_items_renderer:
3848 continue
3849 break
3850
3851 @staticmethod
3852 def _extract_selected_tab(tabs):
3853 for tab in tabs:
3854 renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
3855 if renderer.get('selected') is True:
3856 return renderer
3857 else:
3858 raise ExtractorError('Unable to find selected tab')
3859
3860 @classmethod
3861 def _extract_uploader(cls, data):
3862 uploader = {}
3863 renderer = cls._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
3864 owner = try_get(
3865 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3866 if owner:
3867 uploader['uploader'] = owner.get('text')
3868 uploader['uploader_id'] = try_get(
3869 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3870 uploader['uploader_url'] = urljoin(
3871 'https://www.youtube.com/',
3872 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3873 return {k: v for k, v in uploader.items() if v is not None}
3874
3875 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
3876 playlist_id = title = description = channel_url = channel_name = channel_id = None
3877 thumbnails_list = []
3878 tags = []
3879
3880 selected_tab = self._extract_selected_tab(tabs)
3881 renderer = try_get(
3882 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3883 if renderer:
3884 channel_name = renderer.get('title')
3885 channel_url = renderer.get('channelUrl')
3886 channel_id = renderer.get('externalId')
3887 else:
3888 renderer = try_get(
3889 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3890
3891 if renderer:
3892 title = renderer.get('title')
3893 description = renderer.get('description', '')
3894 playlist_id = channel_id
3895 tags = renderer.get('keywords', '').split()
3896 thumbnails_list = (
3897 try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
3898 or try_get(
3899 self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer'),
3900 lambda x: x['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
3901 list)
3902 or [])
3903
3904 thumbnails = []
3905 for t in thumbnails_list:
3906 if not isinstance(t, dict):
3907 continue
3908 thumbnail_url = url_or_none(t.get('url'))
3909 if not thumbnail_url:
3910 continue
3911 thumbnails.append({
3912 'url': thumbnail_url,
3913 'width': int_or_none(t.get('width')),
3914 'height': int_or_none(t.get('height')),
3915 })
3916 if playlist_id is None:
3917 playlist_id = item_id
3918 if title is None:
3919 title = (
3920 try_get(data, lambda x: x['header']['hashtagHeaderRenderer']['hashtag']['simpleText'])
3921 or playlist_id)
3922 title += format_field(selected_tab, 'title', ' - %s')
3923 title += format_field(selected_tab, 'expandedText', ' - %s')
3924 metadata = {
3925 'playlist_id': playlist_id,
3926 'playlist_title': title,
3927 'playlist_description': description,
3928 'uploader': channel_name,
3929 'uploader_id': channel_id,
3930 'uploader_url': channel_url,
3931 'thumbnails': thumbnails,
3932 'tags': tags,
3933 }
3934 availability = self._extract_availability(data)
3935 if availability:
3936 metadata['availability'] = availability
3937 if not channel_id:
3938 metadata.update(self._extract_uploader(data))
3939 metadata.update({
3940 'channel': metadata['uploader'],
3941 'channel_id': metadata['uploader_id'],
3942 'channel_url': metadata['uploader_url']})
3943 return self.playlist_result(
3944 self._entries(
3945 selected_tab, playlist_id, ytcfg,
3946 self._extract_account_syncid(ytcfg, data),
3947 self._extract_visitor_data(data, ytcfg)),
3948 **metadata)
3949
3950 def _extract_mix_playlist(self, playlist, playlist_id, data, ytcfg):
3951 first_id = last_id = response = None
3952 for page_num in itertools.count(1):
3953 videos = list(self._playlist_entries(playlist))
3954 if not videos:
3955 return
3956 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
3957 if start >= len(videos):
3958 return
3959 for video in videos[start:]:
3960 if video['id'] == first_id:
3961 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
3962 return
3963 yield video
3964 first_id = first_id or videos[0]['id']
3965 last_id = videos[-1]['id']
3966 watch_endpoint = try_get(
3967 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
3968 headers = self.generate_api_headers(
3969 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
3970 visitor_data=self._extract_visitor_data(response, data, ytcfg))
3971 query = {
3972 'playlistId': playlist_id,
3973 'videoId': watch_endpoint.get('videoId') or last_id,
3974 'index': watch_endpoint.get('index') or len(videos),
3975 'params': watch_endpoint.get('params') or 'OAE%3D'
3976 }
3977 response = self._extract_response(
3978 item_id='%s page %d' % (playlist_id, page_num),
3979 query=query, ep='next', headers=headers, ytcfg=ytcfg,
3980 check_get_keys='contents'
3981 )
3982 playlist = try_get(
3983 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3984
3985 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
3986 title = playlist.get('title') or try_get(
3987 data, lambda x: x['titleText']['simpleText'], compat_str)
3988 playlist_id = playlist.get('playlistId') or item_id
3989
3990 # Delegating everything except mix playlists to regular tab-based playlist URL
3991 playlist_url = urljoin(url, try_get(
3992 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3993 compat_str))
3994 if playlist_url and playlist_url != url:
3995 return self.url_result(
3996 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3997 video_title=title)
3998
3999 return self.playlist_result(
4000 self._extract_mix_playlist(playlist, playlist_id, data, ytcfg),
4001 playlist_id=playlist_id, playlist_title=title)
4002
4003 def _extract_availability(self, data):
4004 """
4005 Gets the availability of a given playlist/tab.
4006 Note: Unless YouTube tells us explicitly, we do not assume it is public
4007 @param data: response
4008 """
4009 is_private = is_unlisted = None
4010 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4011 badge_labels = self._extract_badges(renderer)
4012
4013 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4014 privacy_dropdown_entries = try_get(
4015 renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
4016 for renderer_dict in privacy_dropdown_entries:
4017 is_selected = try_get(
4018 renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
4019 if not is_selected:
4020 continue
4021 label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label'))
4022 if label:
4023 badge_labels.add(label.lower())
4024 break
4025
4026 for badge_label in badge_labels:
4027 if badge_label == 'unlisted':
4028 is_unlisted = True
4029 elif badge_label == 'private':
4030 is_private = True
4031 elif badge_label == 'public':
4032 is_unlisted = is_private = False
4033 return self._availability(is_private, False, False, False, is_unlisted)
4034
4035 @staticmethod
4036 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4037 sidebar_renderer = try_get(
4038 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4039 for item in sidebar_renderer:
4040 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4041 if renderer:
4042 return renderer
4043
4044 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4045 """
4046 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
4047 """
4048 browse_id = params = None
4049 renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4050 if not renderer:
4051 return
4052 menu_renderer = try_get(
4053 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
4054 for menu_item in menu_renderer:
4055 if not isinstance(menu_item, dict):
4056 continue
4057 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
4058 text = try_get(
4059 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
4060 if not text or text.lower() != 'show unavailable videos':
4061 continue
4062 browse_endpoint = try_get(
4063 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
4064 browse_id = browse_endpoint.get('browseId')
4065 params = browse_endpoint.get('params')
4066 break
4067
4068 headers = self.generate_api_headers(
4069 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4070 visitor_data=self._extract_visitor_data(data, ytcfg))
4071 query = {
4072 'params': params or 'wgYCCAA=',
4073 'browseId': browse_id or 'VL%s' % item_id
4074 }
4075 return self._extract_response(
4076 item_id=item_id, headers=headers, query=query,
4077 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4078 note='Downloading API JSON with unavailable videos')
4079
4080 def _extract_webpage(self, url, item_id, fatal=True):
4081 retries = self.get_param('extractor_retries', 3)
4082 count = -1
4083 webpage = data = last_error = None
4084 while count < retries:
4085 count += 1
4086 # Sometimes youtube returns a webpage with incomplete ytInitialData
4087 # See: https://github.com/yt-dlp/yt-dlp/issues/116
4088 if last_error:
4089 self.report_warning('%s. Retrying ...' % last_error)
4090 try:
4091 webpage = self._download_webpage(
4092 url, item_id,
4093 note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',))
4094 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4095 except ExtractorError as e:
4096 if isinstance(e.cause, network_exceptions):
4097 if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
4098 last_error = error_to_compat_str(e.cause or e.msg)
4099 if count < retries:
4100 continue
4101 if fatal:
4102 raise
4103 self.report_warning(error_to_compat_str(e))
4104 break
4105 else:
4106 try:
4107 self._extract_and_report_alerts(data)
4108 except ExtractorError as e:
4109 if fatal:
4110 raise
4111 self.report_warning(error_to_compat_str(e))
4112 break
4113
4114 if dict_get(data, ('contents', 'currentVideoEndpoint')):
4115 break
4116
4117 last_error = 'Incomplete yt initial data received'
4118 if count >= retries:
4119 if fatal:
4120 raise ExtractorError(last_error)
4121 self.report_warning(last_error)
4122 break
4123
4124 return webpage, data
4125
4126 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4127 data = None
4128 if 'webpage' not in self._configuration_arg('skip'):
4129 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4130 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4131 if not data:
4132 if not ytcfg and self.is_authenticated:
4133 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.'
4134 if 'authcheck' not in self._configuration_arg('skip') and fatal:
4135 raise ExtractorError(
4136 msg + ' If you are not downloading private content, or your cookies are only for the first account and channel,'
4137 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4138 expected=True)
4139 self.report_warning(msg, only_once=True)
4140 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
4141 return data, ytcfg
4142
4143 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
4144 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
4145 resolve_response = self._extract_response(
4146 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
4147 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
4148 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
4149 for ep_key, ep in endpoints.items():
4150 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
4151 if params:
4152 return self._extract_response(
4153 item_id=item_id, query=params, ep=ep, headers=headers,
4154 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
4155 check_get_keys=('contents', 'currentVideoEndpoint'))
4156 err_note = 'Failed to resolve url (does the playlist exist?)'
4157 if fatal:
4158 raise ExtractorError(err_note, expected=True)
4159 self.report_warning(err_note, item_id)
4160
4161 @staticmethod
4162 def _smuggle_data(entries, data):
4163 for entry in entries:
4164 if data:
4165 entry['url'] = smuggle_url(entry['url'], data)
4166 yield entry
4167
4168 def _real_extract(self, url):
4169 url, smuggled_data = unsmuggle_url(url, {})
4170 if self.is_music_url(url):
4171 smuggled_data['is_music_url'] = True
4172 info_dict = self.__real_extract(url, smuggled_data)
4173 if info_dict.get('entries'):
4174 info_dict['entries'] = self._smuggle_data(info_dict['entries'], smuggled_data)
4175 return info_dict
4176
4177 _url_re = re.compile(r'(?P<pre>%s)(?(channel_type)(?P<tab>/\w+))?(?P<post>.*)$' % _VALID_URL)
4178
4179 def __real_extract(self, url, smuggled_data):
4180 item_id = self._match_id(url)
4181 url = compat_urlparse.urlunparse(
4182 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
4183 compat_opts = self.get_param('compat_opts', [])
4184
4185 def get_mobj(url):
4186 mobj = self._url_re.match(url).groupdict()
4187 mobj.update((k, '') for k, v in mobj.items() if v is None)
4188 return mobj
4189
4190 mobj = get_mobj(url)
4191 # Youtube returns incomplete data if tabname is not lower case
4192 pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel']
4193 if is_channel:
4194 if smuggled_data.get('is_music_url'):
4195 if item_id[:2] == 'VL':
4196 # Youtube music VL channels have an equivalent playlist
4197 item_id = item_id[2:]
4198 pre, tab, post, is_channel = 'https://www.youtube.com/playlist?list=%s' % item_id, '', '', False
4199 elif item_id[:2] == 'MP':
4200 # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
4201 mdata = self._extract_tab_endpoint(
4202 'https://music.youtube.com/channel/%s' % item_id, item_id, default_client='web_music')
4203 murl = traverse_obj(
4204 mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'), get_all=False, expected_type=compat_str)
4205 if not murl:
4206 raise ExtractorError('Failed to resolve album to playlist.')
4207 return self.url_result(murl, ie=YoutubeTabIE.ie_key())
4208 elif mobj['channel_type'] == 'browse':
4209 # Youtube music /browse/ should be changed to /channel/
4210 pre = 'https://www.youtube.com/channel/%s' % item_id
4211 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
4212 # Home URLs should redirect to /videos/
4213 self.report_warning(
4214 'A channel/user page was given. All the channel\'s videos will be downloaded. '
4215 'To download only the videos in the home page, add a "/featured" to the URL')
4216 tab = '/videos'
4217
4218 url = ''.join((pre, tab, post))
4219 mobj = get_mobj(url)
4220
4221 # Handle both video/playlist URLs
4222 qs = parse_qs(url)
4223 video_id = qs.get('v', [None])[0]
4224 playlist_id = qs.get('list', [None])[0]
4225
4226 if not video_id and mobj['not_channel'].startswith('watch'):
4227 if not playlist_id:
4228 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
4229 raise ExtractorError('Unable to recognize tab page')
4230 # Common mistake: https://www.youtube.com/watch?list=playlist_id
4231 self.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
4232 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
4233 mobj = get_mobj(url)
4234
4235 if video_id and playlist_id:
4236 if self.get_param('noplaylist'):
4237 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4238 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', ie=YoutubeIE.ie_key(), video_id=video_id)
4239 self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
4240
4241 data, ytcfg = self._extract_data(url, item_id)
4242
4243 tabs = try_get(
4244 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
4245 if tabs:
4246 selected_tab = self._extract_selected_tab(tabs)
4247 tab_name = selected_tab.get('title', '')
4248 if 'no-youtube-channel-redirect' not in compat_opts:
4249 if mobj['tab'] == '/live':
4250 # Live tab should have redirected to the video
4251 raise ExtractorError('The channel is not currently live', expected=True)
4252 if mobj['tab'] == '/videos' and tab_name.lower() != mobj['tab'][1:]:
4253 if not mobj['not_channel'] and item_id[:2] == 'UC':
4254 # Topic channels don't have /videos. Use the equivalent playlist instead
4255 self.report_warning('The URL does not have a %s tab. Trying to redirect to playlist UU%s instead' % (mobj['tab'][1:], item_id[2:]))
4256 pl_id = 'UU%s' % item_id[2:]
4257 pl_url = 'https://www.youtube.com/playlist?list=%s%s' % (pl_id, mobj['post'])
4258 try:
4259 data, ytcfg, item_id, url = *self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True), pl_id, pl_url
4260 except ExtractorError:
4261 self.report_warning('The playlist gave error. Falling back to channel URL')
4262 else:
4263 self.report_warning('The URL does not have a %s tab. %s is being downloaded instead' % (mobj['tab'][1:], tab_name))
4264
4265 self.write_debug('Final URL: %s' % url)
4266
4267 # YouTube sometimes provides a button to reload playlist with unavailable videos.
4268 if 'no-youtube-unavailable-videos' not in compat_opts:
4269 data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data
4270 self._extract_and_report_alerts(data, only_once=True)
4271 tabs = try_get(
4272 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
4273 if tabs:
4274 return self._extract_from_tabs(item_id, ytcfg, data, tabs)
4275
4276 playlist = try_get(
4277 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4278 if playlist:
4279 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
4280
4281 video_id = try_get(
4282 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
4283 compat_str) or video_id
4284 if video_id:
4285 if mobj['tab'] != '/live': # live tab is expected to redirect to video
4286 self.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
4287 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', ie=YoutubeIE.ie_key(), video_id=video_id)
4288
4289 raise ExtractorError('Unable to recognize tab page')
4290
4291
4292 class YoutubePlaylistIE(InfoExtractor):
4293 IE_DESC = 'YouTube playlists'
4294 _VALID_URL = r'''(?x)(?:
4295 (?:https?://)?
4296 (?:\w+\.)?
4297 (?:
4298 (?:
4299 youtube(?:kids)?\.com|
4300 %(invidious)s
4301 )
4302 /.*?\?.*?\blist=
4303 )?
4304 (?P<id>%(playlist_id)s)
4305 )''' % {
4306 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
4307 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
4308 }
4309 IE_NAME = 'youtube:playlist'
4310 _TESTS = [{
4311 'note': 'issue #673',
4312 'url': 'PLBB231211A4F62143',
4313 'info_dict': {
4314 'title': '[OLD]Team Fortress 2 (Class-based LP)',
4315 'id': 'PLBB231211A4F62143',
4316 'uploader': 'Wickydoo',
4317 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
4318 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
4319 },
4320 'playlist_mincount': 29,
4321 }, {
4322 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
4323 'info_dict': {
4324 'title': 'YDL_safe_search',
4325 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
4326 },
4327 'playlist_count': 2,
4328 'skip': 'This playlist is private',
4329 }, {
4330 'note': 'embedded',
4331 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
4332 'playlist_count': 4,
4333 'info_dict': {
4334 'title': 'JODA15',
4335 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
4336 'uploader': 'milan',
4337 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
4338 }
4339 }, {
4340 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
4341 'playlist_mincount': 654,
4342 'info_dict': {
4343 'title': '2018 Chinese New Singles (11/6 updated)',
4344 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
4345 'uploader': 'LBK',
4346 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
4347 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
4348 }
4349 }, {
4350 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
4351 'only_matching': True,
4352 }, {
4353 # music album playlist
4354 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
4355 'only_matching': True,
4356 }]
4357
4358 @classmethod
4359 def suitable(cls, url):
4360 if YoutubeTabIE.suitable(url):
4361 return False
4362 from ..utils import parse_qs
4363 qs = parse_qs(url)
4364 if qs.get('v', [None])[0]:
4365 return False
4366 return super(YoutubePlaylistIE, cls).suitable(url)
4367
4368 def _real_extract(self, url):
4369 playlist_id = self._match_id(url)
4370 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
4371 url = update_url_query(
4372 'https://www.youtube.com/playlist',
4373 parse_qs(url) or {'list': playlist_id})
4374 if is_music_url:
4375 url = smuggle_url(url, {'is_music_url': True})
4376 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4377
4378
4379 class YoutubeYtBeIE(InfoExtractor):
4380 IE_DESC = 'youtu.be'
4381 _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}
4382 _TESTS = [{
4383 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
4384 'info_dict': {
4385 'id': 'yeWKywCrFtk',
4386 'ext': 'mp4',
4387 'title': 'Small Scale Baler and Braiding Rugs',
4388 'uploader': 'Backus-Page House Museum',
4389 'uploader_id': 'backuspagemuseum',
4390 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
4391 'upload_date': '20161008',
4392 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
4393 'categories': ['Nonprofits & Activism'],
4394 'tags': list,
4395 'like_count': int,
4396 'dislike_count': int,
4397 },
4398 'params': {
4399 'noplaylist': True,
4400 'skip_download': True,
4401 },
4402 }, {
4403 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
4404 'only_matching': True,
4405 }]
4406
4407 def _real_extract(self, url):
4408 mobj = self._match_valid_url(url)
4409 video_id = mobj.group('id')
4410 playlist_id = mobj.group('playlist_id')
4411 return self.url_result(
4412 update_url_query('https://www.youtube.com/watch', {
4413 'v': video_id,
4414 'list': playlist_id,
4415 'feature': 'youtu.be',
4416 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4417
4418
4419 class YoutubeYtUserIE(InfoExtractor):
4420 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
4421 _VALID_URL = r'ytuser:(?P<id>.+)'
4422 _TESTS = [{
4423 'url': 'ytuser:phihag',
4424 'only_matching': True,
4425 }]
4426
4427 def _real_extract(self, url):
4428 user_id = self._match_id(url)
4429 return self.url_result(
4430 'https://www.youtube.com/user/%s' % user_id,
4431 ie=YoutubeTabIE.ie_key(), video_id=user_id)
4432
4433
4434 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
4435 IE_NAME = 'youtube:favorites'
4436 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
4437 _VALID_URL = r':ytfav(?:ou?rite)?s?'
4438 _LOGIN_REQUIRED = True
4439 _TESTS = [{
4440 'url': ':ytfav',
4441 'only_matching': True,
4442 }, {
4443 'url': ':ytfavorites',
4444 'only_matching': True,
4445 }]
4446
4447 def _real_extract(self, url):
4448 return self.url_result(
4449 'https://www.youtube.com/playlist?list=LL',
4450 ie=YoutubeTabIE.ie_key())
4451
4452
4453 class YoutubeSearchIE(SearchInfoExtractor, YoutubeTabIE):
4454 IE_DESC = 'YouTube searches'
4455 IE_NAME = 'youtube:search'
4456 _SEARCH_KEY = 'ytsearch'
4457 _SEARCH_PARAMS = None
4458 _TESTS = []
4459
4460 def _search_results(self, query):
4461 data = {'query': query}
4462 if self._SEARCH_PARAMS:
4463 data['params'] = self._SEARCH_PARAMS
4464 continuation = {}
4465 for page_num in itertools.count(1):
4466 data.update(continuation)
4467 search = self._extract_response(
4468 item_id='query "%s" page %s' % (query, page_num), ep='search', query=data,
4469 check_get_keys=('contents', 'onResponseReceivedCommands')
4470 )
4471 if not search:
4472 break
4473 slr_contents = try_get(
4474 search,
4475 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
4476 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
4477 list)
4478 if not slr_contents:
4479 break
4480
4481 # Youtube sometimes adds promoted content to searches,
4482 # changing the index location of videos and token.
4483 # So we search through all entries till we find them.
4484 continuation = None
4485 for slr_content in slr_contents:
4486 if not continuation:
4487 continuation = self._extract_continuation({'contents': [slr_content]})
4488
4489 isr_contents = try_get(
4490 slr_content,
4491 lambda x: x['itemSectionRenderer']['contents'],
4492 list)
4493 if not isr_contents:
4494 continue
4495 for content in isr_contents:
4496 if not isinstance(content, dict):
4497 continue
4498 video = content.get('videoRenderer')
4499 if not isinstance(video, dict):
4500 continue
4501 video_id = video.get('videoId')
4502 if not video_id:
4503 continue
4504
4505 yield self._extract_video(video)
4506
4507 if not continuation:
4508 break
4509
4510
4511 class YoutubeSearchDateIE(YoutubeSearchIE):
4512 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
4513 _SEARCH_KEY = 'ytsearchdate'
4514 IE_DESC = 'YouTube searches, newest videos first'
4515 _SEARCH_PARAMS = 'CAI%3D'
4516
4517
4518 class YoutubeSearchURLIE(YoutubeSearchIE):
4519 IE_DESC = 'YouTube search URLs with sorting and filter support'
4520 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
4521 _SEARCH_KEY = None
4522 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
4523 # _MAX_RESULTS = 100
4524 _TESTS = [{
4525 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
4526 'playlist_mincount': 5,
4527 'info_dict': {
4528 'id': 'youtube-dl test video',
4529 'title': 'youtube-dl test video',
4530 }
4531 }, {
4532 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
4533 'only_matching': True,
4534 }]
4535
4536 @classmethod
4537 def _make_valid_url(cls):
4538 return cls._VALID_URL
4539
4540 def _real_extract(self, url):
4541 qs = parse_qs(url)
4542 query = (qs.get('search_query') or qs.get('q'))[0]
4543 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
4544 return self._get_n_results(query, self._MAX_RESULTS)
4545
4546
4547 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
4548 """
4549 Base class for feed extractors
4550 Subclasses must define the _FEED_NAME property.
4551 """
4552 _LOGIN_REQUIRED = True
4553 _TESTS = []
4554
4555 @property
4556 def IE_NAME(self):
4557 return 'youtube:%s' % self._FEED_NAME
4558
4559 def _real_extract(self, url):
4560 return self.url_result(
4561 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
4562 ie=YoutubeTabIE.ie_key())
4563
4564
4565 class YoutubeWatchLaterIE(InfoExtractor):
4566 IE_NAME = 'youtube:watchlater'
4567 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
4568 _VALID_URL = r':ytwatchlater'
4569 _TESTS = [{
4570 'url': ':ytwatchlater',
4571 'only_matching': True,
4572 }]
4573
4574 def _real_extract(self, url):
4575 return self.url_result(
4576 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
4577
4578
4579 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
4580 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
4581 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
4582 _FEED_NAME = 'recommended'
4583 _LOGIN_REQUIRED = False
4584 _TESTS = [{
4585 'url': ':ytrec',
4586 'only_matching': True,
4587 }, {
4588 'url': ':ytrecommended',
4589 'only_matching': True,
4590 }, {
4591 'url': 'https://youtube.com',
4592 'only_matching': True,
4593 }]
4594
4595
4596 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
4597 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
4598 _VALID_URL = r':ytsub(?:scription)?s?'
4599 _FEED_NAME = 'subscriptions'
4600 _TESTS = [{
4601 'url': ':ytsubs',
4602 'only_matching': True,
4603 }, {
4604 'url': ':ytsubscriptions',
4605 'only_matching': True,
4606 }]
4607
4608
4609 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
4610 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
4611 _VALID_URL = r':ythis(?:tory)?'
4612 _FEED_NAME = 'history'
4613 _TESTS = [{
4614 'url': ':ythistory',
4615 'only_matching': True,
4616 }]
4617
4618
4619 class YoutubeTruncatedURLIE(InfoExtractor):
4620 IE_NAME = 'youtube:truncated_url'
4621 IE_DESC = False # Do not list
4622 _VALID_URL = r'''(?x)
4623 (?:https?://)?
4624 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
4625 (?:watch\?(?:
4626 feature=[a-z_]+|
4627 annotation_id=annotation_[^&]+|
4628 x-yt-cl=[0-9]+|
4629 hl=[^&]*|
4630 t=[0-9]+
4631 )?
4632 |
4633 attribution_link\?a=[^&]+
4634 )
4635 $
4636 '''
4637
4638 _TESTS = [{
4639 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
4640 'only_matching': True,
4641 }, {
4642 'url': 'https://www.youtube.com/watch?',
4643 'only_matching': True,
4644 }, {
4645 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
4646 'only_matching': True,
4647 }, {
4648 'url': 'https://www.youtube.com/watch?feature=foo',
4649 'only_matching': True,
4650 }, {
4651 'url': 'https://www.youtube.com/watch?hl=en-GB',
4652 'only_matching': True,
4653 }, {
4654 'url': 'https://www.youtube.com/watch?t=2372',
4655 'only_matching': True,
4656 }]
4657
4658 def _real_extract(self, url):
4659 raise ExtractorError(
4660 'Did you forget to quote the URL? Remember that & is a meta '
4661 'character in most shells, so you want to put the URL in quotes, '
4662 'like youtube-dl '
4663 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
4664 ' or simply youtube-dl BaW_jenozKc .',
4665 expected=True)
4666
4667
4668 class YoutubeClipIE(InfoExtractor):
4669 IE_NAME = 'youtube:clip'
4670 IE_DESC = False # Do not list
4671 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/'
4672
4673 def _real_extract(self, url):
4674 self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead')
4675 return self.url_result(url, 'Generic')
4676
4677
4678 class YoutubeTruncatedIDIE(InfoExtractor):
4679 IE_NAME = 'youtube:truncated_id'
4680 IE_DESC = False # Do not list
4681 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
4682
4683 _TESTS = [{
4684 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
4685 'only_matching': True,
4686 }]
4687
4688 def _real_extract(self, url):
4689 video_id = self._match_id(url)
4690 raise ExtractorError(
4691 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
4692 expected=True)