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