]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
projects / yt-dlp.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
Deprecate support for python versions < 3.6
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import calendar
6 import hashlib
7 import itertools
8 import json
9 import os.path
10 import random
11 import re
12 import time
13 import traceback
14
15 from .common import InfoExtractor, SearchInfoExtractor
16 from ..compat import (
17 compat_chr,
18 compat_HTTPError,
19 compat_parse_qs,
20 compat_str,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
24 compat_urlparse,
25 )
26 from ..jsinterp import JSInterpreter
27 from ..utils import (
28 bool_or_none,
29 clean_html,
30 dict_get,
31 datetime_from_str,
32 error_to_compat_str,
33 ExtractorError,
34 format_field,
35 float_or_none,
36 int_or_none,
37 mimetype2ext,
38 parse_codecs,
39 parse_duration,
40 qualities,
41 remove_start,
42 smuggle_url,
43 str_or_none,
44 str_to_int,
45 try_get,
46 unescapeHTML,
47 unified_strdate,
48 unsmuggle_url,
49 update_url_query,
50 url_or_none,
51 urlencode_postdata,
52 urljoin
53 )
54
55
56 def parse_qs(url):
57 return compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
58
59
60 class YoutubeBaseInfoExtractor(InfoExtractor):
61 """Provide base functions for Youtube extractors"""
62 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
63 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
64
65 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
66 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
67 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
68
69 _RESERVED_NAMES = (
70 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|'
71 r'movies|results|shared|hashtag|trending|feed|feeds|oembed|'
72 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
73
74 _NETRC_MACHINE = 'youtube'
75 # If True it will raise an error if no login info is provided
76 _LOGIN_REQUIRED = False
77
78 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
79
80 def _login(self):
81 """
82 Attempt to log in to YouTube.
83 True is returned if successful or skipped.
84 False is returned if login failed.
85
86 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
87 """
88 username, password = self._get_login_info()
89 # No authentication to be performed
90 if username is None:
91 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
92 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
93 # if self._downloader.params.get('cookiefile'): # TODO remove 'and False' later - too many people using outdated cookies and open issues, remind them.
94 # self.to_screen('[Cookies] Reminder - Make sure to always use up to date cookies!')
95 return True
96
97 login_page = self._download_webpage(
98 self._LOGIN_URL, None,
99 note='Downloading login page',
100 errnote='unable to fetch login page', fatal=False)
101 if login_page is False:
102 return
103
104 login_form = self._hidden_inputs(login_page)
105
106 def req(url, f_req, note, errnote):
107 data = login_form.copy()
108 data.update({
109 'pstMsg': 1,
110 'checkConnection': 'youtube',
111 'checkedDomains': 'youtube',
112 'hl': 'en',
113 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
114 'f.req': json.dumps(f_req),
115 'flowName': 'GlifWebSignIn',
116 'flowEntry': 'ServiceLogin',
117 # TODO: reverse actual botguard identifier generation algo
118 'bgRequest': '["identifier",""]',
119 })
120 return self._download_json(
121 url, None, note=note, errnote=errnote,
122 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
123 fatal=False,
124 data=urlencode_postdata(data), headers={
125 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
126 'Google-Accounts-XSRF': 1,
127 })
128
129 def warn(message):
130 self.report_warning(message)
131
132 lookup_req = [
133 username,
134 None, [], None, 'US', None, None, 2, False, True,
135 [
136 None, None,
137 [2, 1, None, 1,
138 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
139 None, [], 4],
140 1, [None, None, []], None, None, None, True
141 ],
142 username,
143 ]
144
145 lookup_results = req(
146 self._LOOKUP_URL, lookup_req,
147 'Looking up account info', 'Unable to look up account info')
148
149 if lookup_results is False:
150 return False
151
152 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
153 if not user_hash:
154 warn('Unable to extract user hash')
155 return False
156
157 challenge_req = [
158 user_hash,
159 None, 1, None, [1, None, None, None, [password, None, True]],
160 [
161 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
162 1, [None, None, []], None, None, None, True
163 ]]
164
165 challenge_results = req(
166 self._CHALLENGE_URL, challenge_req,
167 'Logging in', 'Unable to log in')
168
169 if challenge_results is False:
170 return
171
172 login_res = try_get(challenge_results, lambda x: x[0][5], list)
173 if login_res:
174 login_msg = try_get(login_res, lambda x: x[5], compat_str)
175 warn(
176 'Unable to login: %s' % 'Invalid password'
177 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
178 return False
179
180 res = try_get(challenge_results, lambda x: x[0][-1], list)
181 if not res:
182 warn('Unable to extract result entry')
183 return False
184
185 login_challenge = try_get(res, lambda x: x[0][0], list)
186 if login_challenge:
187 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
188 if challenge_str == 'TWO_STEP_VERIFICATION':
189 # SEND_SUCCESS - TFA code has been successfully sent to phone
190 # QUOTA_EXCEEDED - reached the limit of TFA codes
191 status = try_get(login_challenge, lambda x: x[5], compat_str)
192 if status == 'QUOTA_EXCEEDED':
193 warn('Exceeded the limit of TFA codes, try later')
194 return False
195
196 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
197 if not tl:
198 warn('Unable to extract TL')
199 return False
200
201 tfa_code = self._get_tfa_info('2-step verification code')
202
203 if not tfa_code:
204 warn(
205 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
206 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
207 return False
208
209 tfa_code = remove_start(tfa_code, 'G-')
210
211 tfa_req = [
212 user_hash, None, 2, None,
213 [
214 9, None, None, None, None, None, None, None,
215 [None, tfa_code, True, 2]
216 ]]
217
218 tfa_results = req(
219 self._TFA_URL.format(tl), tfa_req,
220 'Submitting TFA code', 'Unable to submit TFA code')
221
222 if tfa_results is False:
223 return False
224
225 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
226 if tfa_res:
227 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
228 warn(
229 'Unable to finish TFA: %s' % 'Invalid TFA code'
230 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
231 return False
232
233 check_cookie_url = try_get(
234 tfa_results, lambda x: x[0][-1][2], compat_str)
235 else:
236 CHALLENGES = {
237 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
238 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
239 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
240 }
241 challenge = CHALLENGES.get(
242 challenge_str,
243 '%s returned error %s.' % (self.IE_NAME, challenge_str))
244 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
245 return False
246 else:
247 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
248
249 if not check_cookie_url:
250 warn('Unable to extract CheckCookie URL')
251 return False
252
253 check_cookie_results = self._download_webpage(
254 check_cookie_url, None, 'Checking cookie', fatal=False)
255
256 if check_cookie_results is False:
257 return False
258
259 if 'https://myaccount.google.com/' not in check_cookie_results:
260 warn('Unable to log in')
261 return False
262
263 return True
264
265 def _initialize_consent(self):
266 cookies = self._get_cookies('https://www.youtube.com/')
267 if cookies.get('__Secure-3PSID'):
268 return
269 consent_id = None
270 consent = cookies.get('CONSENT')
271 if consent:
272 if 'YES' in consent.value:
273 return
274 consent_id = self._search_regex(
275 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
276 if not consent_id:
277 consent_id = random.randint(100, 999)
278 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
279
280 def _real_initialize(self):
281 self._initialize_consent()
282 if self._downloader is None:
283 return
284 if not self._login():
285 return
286
287 _YT_WEB_CLIENT_VERSION = '2.20210407.08.00'
288 _YT_INNERTUBE_API_KEY = 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'
289 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
290 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
291 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
292
293 def _generate_sapisidhash_header(self):
294 sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
295 if sapisid_cookie is None:
296 return
297 time_now = round(time.time())
298 sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
299 return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
300
301 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
302 note='Downloading API JSON', errnote='Unable to download API page',
303 context=None, api_key=None):
304
305 data = {'context': context} if context else {'context': self._extract_context()}
306 data.update(query)
307 real_headers = self._generate_api_headers()
308 real_headers.update({'content-type': 'application/json'})
309 if headers:
310 real_headers.update(headers)
311 return self._download_json(
312 'https://www.youtube.com/youtubei/v1/%s' % ep,
313 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
314 data=json.dumps(data).encode('utf8'), headers=real_headers,
315 query={'key': api_key or self._extract_api_key()})
316
317 def _extract_api_key(self, ytcfg=None):
318 return try_get(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str) or self._YT_INNERTUBE_API_KEY
319
320 def _extract_yt_initial_data(self, video_id, webpage):
321 return self._parse_json(
322 self._search_regex(
323 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
324 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
325 video_id)
326
327 def _extract_identity_token(self, webpage, item_id):
328 ytcfg = self._extract_ytcfg(item_id, webpage)
329 if ytcfg:
330 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
331 if token:
332 return token
333 return self._search_regex(
334 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
335 'identity token', default=None)
336
337 @staticmethod
338 def _extract_account_syncid(data):
339 """
340 Extract syncId required to download private playlists of secondary channels
341 @param data Either response or ytcfg
342 """
343 sync_ids = (try_get(
344 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
345 lambda x: x['DATASYNC_ID']), compat_str) or '').split("||")
346 if len(sync_ids) >= 2 and sync_ids[1]:
347 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
348 # and just "user_syncid||" for primary channel. We only want the channel_syncid
349 return sync_ids[0]
350 # ytcfg includes channel_syncid if on secondary channel
351 return data.get('DELEGATED_SESSION_ID')
352
353 def _extract_ytcfg(self, video_id, webpage):
354 if not webpage:
355 return {}
356 return self._parse_json(
357 self._search_regex(
358 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
359 default='{}'), video_id, fatal=False) or {}
360
361 def __extract_client_version(self, ytcfg):
362 return try_get(ytcfg, lambda x: x['INNERTUBE_CLIENT_VERSION'], compat_str) or self._YT_WEB_CLIENT_VERSION
363
364 def _extract_context(self, ytcfg=None):
365 context = try_get(ytcfg, lambda x: x['INNERTUBE_CONTEXT'], dict)
366 if context:
367 return context
368
369 # Recreate the client context (required)
370 client_version = self.__extract_client_version(ytcfg)
371 client_name = try_get(ytcfg, lambda x: x['INNERTUBE_CLIENT_NAME'], compat_str) or 'WEB'
372 context = {
373 'client': {
374 'clientName': client_name,
375 'clientVersion': client_version,
376 }
377 }
378 visitor_data = try_get(ytcfg, lambda x: x['VISITOR_DATA'], compat_str)
379 if visitor_data:
380 context['client']['visitorData'] = visitor_data
381 return context
382
383 def _generate_api_headers(self, ytcfg=None, identity_token=None, account_syncid=None, visitor_data=None):
384 headers = {
385 'X-YouTube-Client-Name': '1',
386 'X-YouTube-Client-Version': self.__extract_client_version(ytcfg),
387 }
388 if identity_token:
389 headers['x-youtube-identity-token'] = identity_token
390 if account_syncid:
391 headers['X-Goog-PageId'] = account_syncid
392 headers['X-Goog-AuthUser'] = 0
393 if visitor_data:
394 headers['x-goog-visitor-id'] = visitor_data
395 auth = self._generate_sapisidhash_header()
396 if auth is not None:
397 headers['Authorization'] = auth
398 headers['X-Origin'] = 'https://www.youtube.com'
399 return headers
400
401 def _extract_video(self, renderer):
402 video_id = renderer.get('videoId')
403 title = try_get(
404 renderer,
405 (lambda x: x['title']['runs'][0]['text'],
406 lambda x: x['title']['simpleText']), compat_str)
407 description = try_get(
408 renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'],
409 compat_str)
410 duration = parse_duration(try_get(
411 renderer, lambda x: x['lengthText']['simpleText'], compat_str))
412 view_count_text = try_get(
413 renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
414 view_count = str_to_int(self._search_regex(
415 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
416 'view count', default=None))
417 uploader = try_get(
418 renderer,
419 (lambda x: x['ownerText']['runs'][0]['text'],
420 lambda x: x['shortBylineText']['runs'][0]['text']), compat_str)
421 return {
422 '_type': 'url',
423 'ie_key': YoutubeIE.ie_key(),
424 'id': video_id,
425 'url': video_id,
426 'title': title,
427 'description': description,
428 'duration': duration,
429 'view_count': view_count,
430 'uploader': uploader,
431 }
432
433
434 class YoutubeIE(YoutubeBaseInfoExtractor):
435 IE_DESC = 'YouTube.com'
436 _INVIDIOUS_SITES = (
437 # invidious-redirect websites
438 r'(?:www\.)?redirect\.invidious\.io',
439 r'(?:(?:www|dev)\.)?invidio\.us',
440 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
441 r'(?:www\.)?invidious\.pussthecat\.org',
442 r'(?:www\.)?invidious\.zee\.li',
443 r'(?:(?:www|au)\.)?ytprivate\.com',
444 r'(?:www\.)?invidious\.namazso\.eu',
445 r'(?:www\.)?invidious\.ethibox\.fr',
446 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
447 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
448 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
449 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
450 # youtube-dl invidious instances list
451 r'(?:(?:www|no)\.)?invidiou\.sh',
452 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
453 r'(?:www\.)?invidious\.kabi\.tk',
454 r'(?:www\.)?invidious\.mastodon\.host',
455 r'(?:www\.)?invidious\.zapashcanon\.fr',
456 r'(?:www\.)?invidious\.kavin\.rocks',
457 r'(?:www\.)?invidious\.tinfoil-hat\.net',
458 r'(?:www\.)?invidious\.himiko\.cloud',
459 r'(?:www\.)?invidious\.reallyancient\.tech',
460 r'(?:www\.)?invidious\.tube',
461 r'(?:www\.)?invidiou\.site',
462 r'(?:www\.)?invidious\.site',
463 r'(?:www\.)?invidious\.xyz',
464 r'(?:www\.)?invidious\.nixnet\.xyz',
465 r'(?:www\.)?invidious\.048596\.xyz',
466 r'(?:www\.)?invidious\.drycat\.fr',
467 r'(?:www\.)?inv\.skyn3t\.in',
468 r'(?:www\.)?tube\.poal\.co',
469 r'(?:www\.)?tube\.connect\.cafe',
470 r'(?:www\.)?vid\.wxzm\.sx',
471 r'(?:www\.)?vid\.mint\.lgbt',
472 r'(?:www\.)?vid\.puffyan\.us',
473 r'(?:www\.)?yewtu\.be',
474 r'(?:www\.)?yt\.elukerio\.org',
475 r'(?:www\.)?yt\.lelux\.fi',
476 r'(?:www\.)?invidious\.ggc-project\.de',
477 r'(?:www\.)?yt\.maisputain\.ovh',
478 r'(?:www\.)?ytprivate\.com',
479 r'(?:www\.)?invidious\.13ad\.de',
480 r'(?:www\.)?invidious\.toot\.koeln',
481 r'(?:www\.)?invidious\.fdn\.fr',
482 r'(?:www\.)?watch\.nettohikari\.com',
483 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
484 r'(?:www\.)?qklhadlycap4cnod\.onion',
485 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
486 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
487 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
488 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
489 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
490 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
491 )
492 _VALID_URL = r"""(?x)^
493 (
494 (?:https?://|//) # http(s):// or protocol-independent URL
495 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
496 (?:www\.)?deturl\.com/www\.youtube\.com|
497 (?:www\.)?pwnyoutube\.com|
498 (?:www\.)?hooktube\.com|
499 (?:www\.)?yourepeat\.com|
500 tube\.majestyc\.net|
501 %(invidious)s|
502 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
503 (?:.*?\#/)? # handle anchor (#/) redirect urls
504 (?: # the various things that can precede the ID:
505 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
506 |(?: # or the v= param in all its forms
507 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
508 (?:\?|\#!?) # the params delimiter ? or # or #!
509 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
510 v=
511 )
512 ))
513 |(?:
514 youtu\.be| # just youtu.be/xxxx
515 vid\.plus| # or vid.plus/xxxx
516 zwearz\.com/watch| # or zwearz.com/watch/xxxx
517 %(invidious)s
518 )/
519 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
520 )
521 )? # all until now is optional -> you can pass the naked ID
522 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
523 (?(1).+)? # if we found the ID, everything can follow
524 $""" % {
525 'invidious': '|'.join(_INVIDIOUS_SITES),
526 }
527 _PLAYER_INFO_RE = (
528 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
529 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
530 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
531 )
532 _formats = {
533 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
534 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
535 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
536 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
537 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
538 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
539 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
540 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
541 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
542 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
543 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
544 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
545 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
546 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
547 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
548 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
549 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
550 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
551
552
553 # 3D videos
554 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
555 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
556 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
557 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
558 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
559 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
560 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
561
562 # Apple HTTP Live Streaming
563 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
564 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
565 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
566 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
567 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
568 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
569 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
570 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
571
572 # DASH mp4 video
573 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
574 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
575 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
576 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
577 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
578 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
579 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
580 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
581 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
582 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
583 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
584 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
585
586 # Dash mp4 audio
587 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
588 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
589 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
590 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
591 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
592 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
593 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
594
595 # Dash webm
596 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
597 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
598 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
599 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
600 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
601 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
602 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
603 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
604 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
605 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
606 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
607 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
608 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
609 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
610 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
611 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
612 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
613 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
614 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
615 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
616 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
617 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
618
619 # Dash webm audio
620 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
621 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
622
623 # Dash webm audio with opus inside
624 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
625 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
626 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
627
628 # RTMP (unnamed)
629 '_rtmp': {'protocol': 'rtmp'},
630
631 # av01 video only formats sometimes served with "unknown" codecs
632 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
633 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
634 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
635 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
636 }
637 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
638
639 _GEO_BYPASS = False
640
641 IE_NAME = 'youtube'
642 _TESTS = [
643 {
644 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
645 'info_dict': {
646 'id': 'BaW_jenozKc',
647 'ext': 'mp4',
648 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
649 'uploader': 'Philipp Hagemeister',
650 'uploader_id': 'phihag',
651 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
652 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
653 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
654 'upload_date': '20121002',
655 '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 .',
656 'categories': ['Science & Technology'],
657 'tags': ['youtube-dl'],
658 'duration': 10,
659 'view_count': int,
660 'like_count': int,
661 'dislike_count': int,
662 'start_time': 1,
663 'end_time': 9,
664 }
665 },
666 {
667 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
668 'note': 'Embed-only video (#1746)',
669 'info_dict': {
670 'id': 'yZIXLfi8CZQ',
671 'ext': 'mp4',
672 'upload_date': '20120608',
673 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
674 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
675 'uploader': 'SET India',
676 'uploader_id': 'setindia',
677 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
678 'age_limit': 18,
679 },
680 'skip': 'Private video',
681 },
682 {
683 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
684 'note': 'Use the first video ID in the URL',
685 'info_dict': {
686 'id': 'BaW_jenozKc',
687 'ext': 'mp4',
688 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
689 'uploader': 'Philipp Hagemeister',
690 'uploader_id': 'phihag',
691 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
692 'upload_date': '20121002',
693 '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 .',
694 'categories': ['Science & Technology'],
695 'tags': ['youtube-dl'],
696 'duration': 10,
697 'view_count': int,
698 'like_count': int,
699 'dislike_count': int,
700 },
701 'params': {
702 'skip_download': True,
703 },
704 },
705 {
706 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
707 'note': '256k DASH audio (format 141) via DASH manifest',
708 'info_dict': {
709 'id': 'a9LDPn-MO4I',
710 'ext': 'm4a',
711 'upload_date': '20121002',
712 'uploader_id': '8KVIDEO',
713 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
714 'description': '',
715 'uploader': '8KVIDEO',
716 'title': 'UHDTV TEST 8K VIDEO.mp4'
717 },
718 'params': {
719 'youtube_include_dash_manifest': True,
720 'format': '141',
721 },
722 'skip': 'format 141 not served anymore',
723 },
724 # DASH manifest with encrypted signature
725 {
726 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
727 'info_dict': {
728 'id': 'IB3lcPjvWLA',
729 'ext': 'm4a',
730 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
731 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
732 'duration': 244,
733 'uploader': 'AfrojackVEVO',
734 'uploader_id': 'AfrojackVEVO',
735 'upload_date': '20131011',
736 'abr': 129.495,
737 },
738 'params': {
739 'youtube_include_dash_manifest': True,
740 'format': '141/bestaudio[ext=m4a]',
741 },
742 },
743 # Controversy video
744 {
745 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
746 'info_dict': {
747 'id': 'T4XJQO3qol8',
748 'ext': 'mp4',
749 'duration': 219,
750 'upload_date': '20100909',
751 'uploader': 'Amazing Atheist',
752 'uploader_id': 'TheAmazingAtheist',
753 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
754 'title': 'Burning Everyone\'s Koran',
755 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
756 }
757 },
758 # Normal age-gate video (embed allowed)
759 {
760 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
761 'info_dict': {
762 'id': 'HtVdAasjOgU',
763 'ext': 'mp4',
764 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
765 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
766 'duration': 142,
767 'uploader': 'The Witcher',
768 'uploader_id': 'WitcherGame',
769 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
770 'upload_date': '20140605',
771 'age_limit': 18,
772 },
773 },
774 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
775 # YouTube Red ad is not captured for creator
776 {
777 'url': '__2ABJjxzNo',
778 'info_dict': {
779 'id': '__2ABJjxzNo',
780 'ext': 'mp4',
781 'duration': 266,
782 'upload_date': '20100430',
783 'uploader_id': 'deadmau5',
784 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
785 'creator': 'deadmau5',
786 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
787 'uploader': 'deadmau5',
788 'title': 'Deadmau5 - Some Chords (HD)',
789 'alt_title': 'Some Chords',
790 },
791 'expected_warnings': [
792 'DASH manifest missing',
793 ]
794 },
795 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
796 {
797 'url': 'lqQg6PlCWgI',
798 'info_dict': {
799 'id': 'lqQg6PlCWgI',
800 'ext': 'mp4',
801 'duration': 6085,
802 'upload_date': '20150827',
803 'uploader_id': 'olympic',
804 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
805 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
806 'uploader': 'Olympic',
807 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
808 },
809 'params': {
810 'skip_download': 'requires avconv',
811 }
812 },
813 # Non-square pixels
814 {
815 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
816 'info_dict': {
817 'id': '_b-2C3KPAM0',
818 'ext': 'mp4',
819 'stretched_ratio': 16 / 9.,
820 'duration': 85,
821 'upload_date': '20110310',
822 'uploader_id': 'AllenMeow',
823 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
824 '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',
825 'uploader': '孫ᄋᄅ',
826 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
827 },
828 },
829 # url_encoded_fmt_stream_map is empty string
830 {
831 'url': 'qEJwOuvDf7I',
832 'info_dict': {
833 'id': 'qEJwOuvDf7I',
834 'ext': 'webm',
835 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
836 'description': '',
837 'upload_date': '20150404',
838 'uploader_id': 'spbelect',
839 'uploader': 'Наблюдатели Петербурга',
840 },
841 'params': {
842 'skip_download': 'requires avconv',
843 },
844 'skip': 'This live event has ended.',
845 },
846 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
847 {
848 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
849 'info_dict': {
850 'id': 'FIl7x6_3R5Y',
851 'ext': 'webm',
852 'title': 'md5:7b81415841e02ecd4313668cde88737a',
853 'description': 'md5:116377fd2963b81ec4ce64b542173306',
854 'duration': 220,
855 'upload_date': '20150625',
856 'uploader_id': 'dorappi2000',
857 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
858 'uploader': 'dorappi2000',
859 'formats': 'mincount:31',
860 },
861 'skip': 'not actual anymore',
862 },
863 # DASH manifest with segment_list
864 {
865 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
866 'md5': '8ce563a1d667b599d21064e982ab9e31',
867 'info_dict': {
868 'id': 'CsmdDsKjzN8',
869 'ext': 'mp4',
870 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
871 'uploader': 'Airtek',
872 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
873 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
874 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
875 },
876 'params': {
877 'youtube_include_dash_manifest': True,
878 'format': '135', # bestvideo
879 },
880 'skip': 'This live event has ended.',
881 },
882 {
883 # Multifeed videos (multiple cameras), URL is for Main Camera
884 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
885 'info_dict': {
886 'id': 'jvGDaLqkpTg',
887 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
888 'description': 'md5:e03b909557865076822aa169218d6a5d',
889 },
890 'playlist': [{
891 'info_dict': {
892 'id': 'jvGDaLqkpTg',
893 'ext': 'mp4',
894 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
895 'description': 'md5:e03b909557865076822aa169218d6a5d',
896 'duration': 10643,
897 'upload_date': '20161111',
898 'uploader': 'Team PGP',
899 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
900 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
901 },
902 }, {
903 'info_dict': {
904 'id': '3AKt1R1aDnw',
905 'ext': 'mp4',
906 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
907 'description': 'md5:e03b909557865076822aa169218d6a5d',
908 'duration': 10991,
909 'upload_date': '20161111',
910 'uploader': 'Team PGP',
911 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
912 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
913 },
914 }, {
915 'info_dict': {
916 'id': 'RtAMM00gpVc',
917 'ext': 'mp4',
918 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
919 'description': 'md5:e03b909557865076822aa169218d6a5d',
920 'duration': 10995,
921 'upload_date': '20161111',
922 'uploader': 'Team PGP',
923 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
924 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
925 },
926 }, {
927 'info_dict': {
928 'id': '6N2fdlP3C5U',
929 'ext': 'mp4',
930 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
931 'description': 'md5:e03b909557865076822aa169218d6a5d',
932 'duration': 10990,
933 'upload_date': '20161111',
934 'uploader': 'Team PGP',
935 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
936 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
937 },
938 }],
939 'params': {
940 'skip_download': True,
941 },
942 },
943 {
944 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
945 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
946 'info_dict': {
947 'id': 'gVfLd0zydlo',
948 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
949 },
950 'playlist_count': 2,
951 'skip': 'Not multifeed anymore',
952 },
953 {
954 'url': 'https://vid.plus/FlRa-iH7PGw',
955 'only_matching': True,
956 },
957 {
958 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
959 'only_matching': True,
960 },
961 {
962 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
963 # Also tests cut-off URL expansion in video description (see
964 # https://github.com/ytdl-org/youtube-dl/issues/1892,
965 # https://github.com/ytdl-org/youtube-dl/issues/8164)
966 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
967 'info_dict': {
968 'id': 'lsguqyKfVQg',
969 'ext': 'mp4',
970 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
971 'alt_title': 'Dark Walk - Position Music',
972 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
973 'duration': 133,
974 'upload_date': '20151119',
975 'uploader_id': 'IronSoulElf',
976 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
977 'uploader': 'IronSoulElf',
978 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
979 'track': 'Dark Walk - Position Music',
980 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
981 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
982 },
983 'params': {
984 'skip_download': True,
985 },
986 },
987 {
988 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
989 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
990 'only_matching': True,
991 },
992 {
993 # Video with yt:stretch=17:0
994 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
995 'info_dict': {
996 'id': 'Q39EVAstoRM',
997 'ext': 'mp4',
998 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
999 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1000 'upload_date': '20151107',
1001 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1002 'uploader': 'CH GAMER DROID',
1003 },
1004 'params': {
1005 'skip_download': True,
1006 },
1007 'skip': 'This video does not exist.',
1008 },
1009 {
1010 # Video with incomplete 'yt:stretch=16:'
1011 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1012 'only_matching': True,
1013 },
1014 {
1015 # Video licensed under Creative Commons
1016 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1017 'info_dict': {
1018 'id': 'M4gD1WSo5mA',
1019 'ext': 'mp4',
1020 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1021 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1022 'duration': 721,
1023 'upload_date': '20150127',
1024 'uploader_id': 'BerkmanCenter',
1025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1026 'uploader': 'The Berkman Klein Center for Internet & Society',
1027 'license': 'Creative Commons Attribution license (reuse allowed)',
1028 },
1029 'params': {
1030 'skip_download': True,
1031 },
1032 },
1033 {
1034 # Channel-like uploader_url
1035 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1036 'info_dict': {
1037 'id': 'eQcmzGIKrzg',
1038 'ext': 'mp4',
1039 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1040 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1041 'duration': 4060,
1042 'upload_date': '20151119',
1043 'uploader': 'Bernie Sanders',
1044 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1045 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1046 'license': 'Creative Commons Attribution license (reuse allowed)',
1047 },
1048 'params': {
1049 'skip_download': True,
1050 },
1051 },
1052 {
1053 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1054 'only_matching': True,
1055 },
1056 {
1057 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1058 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1059 'only_matching': True,
1060 },
1061 {
1062 # Rental video preview
1063 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1064 'info_dict': {
1065 'id': 'uGpuVWrhIzE',
1066 'ext': 'mp4',
1067 'title': 'Piku - Trailer',
1068 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1069 'upload_date': '20150811',
1070 'uploader': 'FlixMatrix',
1071 'uploader_id': 'FlixMatrixKaravan',
1072 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1073 'license': 'Standard YouTube License',
1074 },
1075 'params': {
1076 'skip_download': True,
1077 },
1078 'skip': 'This video is not available.',
1079 },
1080 {
1081 # YouTube Red video with episode data
1082 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1083 'info_dict': {
1084 'id': 'iqKdEhx-dD4',
1085 'ext': 'mp4',
1086 'title': 'Isolation - Mind Field (Ep 1)',
1087 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1088 'duration': 2085,
1089 'upload_date': '20170118',
1090 'uploader': 'Vsauce',
1091 'uploader_id': 'Vsauce',
1092 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1093 'series': 'Mind Field',
1094 'season_number': 1,
1095 'episode_number': 1,
1096 },
1097 'params': {
1098 'skip_download': True,
1099 },
1100 'expected_warnings': [
1101 'Skipping DASH manifest',
1102 ],
1103 },
1104 {
1105 # The following content has been identified by the YouTube community
1106 # as inappropriate or offensive to some audiences.
1107 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1108 'info_dict': {
1109 'id': '6SJNVb0GnPI',
1110 'ext': 'mp4',
1111 'title': 'Race Differences in Intelligence',
1112 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1113 'duration': 965,
1114 'upload_date': '20140124',
1115 'uploader': 'New Century Foundation',
1116 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1117 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1118 },
1119 'params': {
1120 'skip_download': True,
1121 },
1122 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1123 },
1124 {
1125 # itag 212
1126 'url': '1t24XAntNCY',
1127 'only_matching': True,
1128 },
1129 {
1130 # geo restricted to JP
1131 'url': 'sJL6WA-aGkQ',
1132 'only_matching': True,
1133 },
1134 {
1135 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1136 'only_matching': True,
1137 },
1138 {
1139 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1140 'only_matching': True,
1141 },
1142 {
1143 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1144 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1145 'only_matching': True,
1146 },
1147 {
1148 # DRM protected
1149 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1150 'only_matching': True,
1151 },
1152 {
1153 # Video with unsupported adaptive stream type formats
1154 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1155 'info_dict': {
1156 'id': 'Z4Vy8R84T1U',
1157 'ext': 'mp4',
1158 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1159 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1160 'duration': 433,
1161 'upload_date': '20130923',
1162 'uploader': 'Amelia Putri Harwita',
1163 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1164 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1165 'formats': 'maxcount:10',
1166 },
1167 'params': {
1168 'skip_download': True,
1169 'youtube_include_dash_manifest': False,
1170 },
1171 'skip': 'not actual anymore',
1172 },
1173 {
1174 # Youtube Music Auto-generated description
1175 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1176 'info_dict': {
1177 'id': 'MgNrAu2pzNs',
1178 'ext': 'mp4',
1179 'title': 'Voyeur Girl',
1180 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1181 'upload_date': '20190312',
1182 'uploader': 'Stephen - Topic',
1183 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1184 'artist': 'Stephen',
1185 'track': 'Voyeur Girl',
1186 'album': 'it\'s too much love to know my dear',
1187 'release_date': '20190313',
1188 'release_year': 2019,
1189 },
1190 'params': {
1191 'skip_download': True,
1192 },
1193 },
1194 {
1195 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1196 'only_matching': True,
1197 },
1198 {
1199 # invalid -> valid video id redirection
1200 'url': 'DJztXj2GPfl',
1201 'info_dict': {
1202 'id': 'DJztXj2GPfk',
1203 'ext': 'mp4',
1204 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1205 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1206 'upload_date': '20090125',
1207 'uploader': 'Prochorowka',
1208 'uploader_id': 'Prochorowka',
1209 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1210 'artist': 'Panjabi MC',
1211 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1212 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1213 },
1214 'params': {
1215 'skip_download': True,
1216 },
1217 'skip': 'Video unavailable',
1218 },
1219 {
1220 # empty description results in an empty string
1221 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1222 'info_dict': {
1223 'id': 'x41yOUIvK2k',
1224 'ext': 'mp4',
1225 'title': 'IMG 3456',
1226 'description': '',
1227 'upload_date': '20170613',
1228 'uploader_id': 'ElevageOrVert',
1229 'uploader': 'ElevageOrVert',
1230 },
1231 'params': {
1232 'skip_download': True,
1233 },
1234 },
1235 {
1236 # with '};' inside yt initial data (see [1])
1237 # see [2] for an example with '};' inside ytInitialPlayerResponse
1238 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1239 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1240 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1241 'info_dict': {
1242 'id': 'CHqg6qOn4no',
1243 'ext': 'mp4',
1244 'title': 'Part 77 Sort a list of simple types in c#',
1245 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1246 'upload_date': '20130831',
1247 'uploader_id': 'kudvenkat',
1248 'uploader': 'kudvenkat',
1249 },
1250 'params': {
1251 'skip_download': True,
1252 },
1253 },
1254 {
1255 # another example of '};' in ytInitialData
1256 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1257 'only_matching': True,
1258 },
1259 {
1260 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1261 'only_matching': True,
1262 },
1263 {
1264 # https://github.com/ytdl-org/youtube-dl/pull/28094
1265 'url': 'OtqTfy26tG0',
1266 'info_dict': {
1267 'id': 'OtqTfy26tG0',
1268 'ext': 'mp4',
1269 'title': 'Burn Out',
1270 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1271 'upload_date': '20141120',
1272 'uploader': 'The Cinematic Orchestra - Topic',
1273 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1274 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1275 'artist': 'The Cinematic Orchestra',
1276 'track': 'Burn Out',
1277 'album': 'Every Day',
1278 'release_data': None,
1279 'release_year': None,
1280 },
1281 'params': {
1282 'skip_download': True,
1283 },
1284 },
1285 {
1286 # controversial video, only works with bpctr when authenticated with cookies
1287 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1288 'only_matching': True,
1289 },
1290 {
1291 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
1292 'url': 'cBvYw8_A0vQ',
1293 'info_dict': {
1294 'id': 'cBvYw8_A0vQ',
1295 'ext': 'mp4',
1296 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
1297 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
1298 'upload_date': '20201120',
1299 'uploader': 'Walk around Japan',
1300 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
1301 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
1302 },
1303 'params': {
1304 'skip_download': True,
1305 },
1306 },
1307 ]
1308
1309 @classmethod
1310 def suitable(cls, url):
1311 # Hack for lazy extractors until more generic solution is implemented
1312 # (see #28780)
1313 from .youtube import parse_qs
1314 qs = parse_qs(url)
1315 if qs.get('list', [None])[0]:
1316 return False
1317 return super(YoutubeIE, cls).suitable(url)
1318
1319 def __init__(self, *args, **kwargs):
1320 super(YoutubeIE, self).__init__(*args, **kwargs)
1321 self._code_cache = {}
1322 self._player_cache = {}
1323
1324 def _signature_cache_id(self, example_sig):
1325 """ Return a string representation of a signature """
1326 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1327
1328 @classmethod
1329 def _extract_player_info(cls, player_url):
1330 for player_re in cls._PLAYER_INFO_RE:
1331 id_m = re.search(player_re, player_url)
1332 if id_m:
1333 break
1334 else:
1335 raise ExtractorError('Cannot identify player %r' % player_url)
1336 return id_m.group('id')
1337
1338 def _extract_signature_function(self, video_id, player_url, example_sig):
1339 player_id = self._extract_player_info(player_url)
1340
1341 # Read from filesystem cache
1342 func_id = 'js_%s_%s' % (
1343 player_id, self._signature_cache_id(example_sig))
1344 assert os.path.basename(func_id) == func_id
1345
1346 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1347 if cache_spec is not None:
1348 return lambda s: ''.join(s[i] for i in cache_spec)
1349
1350 if player_id not in self._code_cache:
1351 self._code_cache[player_id] = self._download_webpage(
1352 player_url, video_id,
1353 note='Downloading player ' + player_id,
1354 errnote='Download of %s failed' % player_url)
1355 code = self._code_cache[player_id]
1356 res = self._parse_sig_js(code)
1357
1358 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1359 cache_res = res(test_string)
1360 cache_spec = [ord(c) for c in cache_res]
1361
1362 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1363 return res
1364
1365 def _print_sig_code(self, func, example_sig):
1366 def gen_sig_code(idxs):
1367 def _genslice(start, end, step):
1368 starts = '' if start == 0 else str(start)
1369 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1370 steps = '' if step == 1 else (':%d' % step)
1371 return 's[%s%s%s]' % (starts, ends, steps)
1372
1373 step = None
1374 # Quelch pyflakes warnings - start will be set when step is set
1375 start = '(Never used)'
1376 for i, prev in zip(idxs[1:], idxs[:-1]):
1377 if step is not None:
1378 if i - prev == step:
1379 continue
1380 yield _genslice(start, prev, step)
1381 step = None
1382 continue
1383 if i - prev in [-1, 1]:
1384 step = i - prev
1385 start = prev
1386 continue
1387 else:
1388 yield 's[%d]' % prev
1389 if step is None:
1390 yield 's[%d]' % i
1391 else:
1392 yield _genslice(start, i, step)
1393
1394 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1395 cache_res = func(test_string)
1396 cache_spec = [ord(c) for c in cache_res]
1397 expr_code = ' + '.join(gen_sig_code(cache_spec))
1398 signature_id_tuple = '(%s)' % (
1399 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1400 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1401 ' return %s\n') % (signature_id_tuple, expr_code)
1402 self.to_screen('Extracted signature function:\n' + code)
1403
1404 def _parse_sig_js(self, jscode):
1405 funcname = self._search_regex(
1406 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1407 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1408 r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
1409 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
1410 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}\(a,\d+\)',
1411 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\(\s*""\s*\)',
1412 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1413 # Obsolete patterns
1414 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1415 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1416 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1417 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1418 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1419 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1420 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1421 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1422 jscode, 'Initial JS player signature function name', group='sig')
1423
1424 jsi = JSInterpreter(jscode)
1425 initial_function = jsi.extract_function(funcname)
1426 return lambda s: initial_function([s])
1427
1428 def _decrypt_signature(self, s, video_id, player_url):
1429 """Turn the encrypted s field into a working signature"""
1430
1431 if player_url is None:
1432 raise ExtractorError('Cannot decrypt signature without player_url')
1433
1434 if player_url.startswith('//'):
1435 player_url = 'https:' + player_url
1436 elif not re.match(r'https?://', player_url):
1437 player_url = compat_urlparse.urljoin(
1438 'https://www.youtube.com', player_url)
1439 try:
1440 player_id = (player_url, self._signature_cache_id(s))
1441 if player_id not in self._player_cache:
1442 func = self._extract_signature_function(
1443 video_id, player_url, s
1444 )
1445 self._player_cache[player_id] = func
1446 func = self._player_cache[player_id]
1447 if self._downloader.params.get('youtube_print_sig_code'):
1448 self._print_sig_code(func, s)
1449 return func(s)
1450 except Exception as e:
1451 tb = traceback.format_exc()
1452 raise ExtractorError(
1453 'Signature extraction failed: ' + tb, cause=e)
1454
1455 def _mark_watched(self, video_id, player_response):
1456 playback_url = url_or_none(try_get(
1457 player_response,
1458 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']))
1459 if not playback_url:
1460 return
1461 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1462 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1463
1464 # cpn generation algorithm is reverse engineered from base.js.
1465 # In fact it works even with dummy cpn.
1466 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1467 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1468
1469 qs.update({
1470 'ver': ['2'],
1471 'cpn': [cpn],
1472 })
1473 playback_url = compat_urlparse.urlunparse(
1474 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1475
1476 self._download_webpage(
1477 playback_url, video_id, 'Marking watched',
1478 'Unable to mark watched', fatal=False)
1479
1480 @staticmethod
1481 def _extract_urls(webpage):
1482 # Embedded YouTube player
1483 entries = [
1484 unescapeHTML(mobj.group('url'))
1485 for mobj in re.finditer(r'''(?x)
1486 (?:
1487 <iframe[^>]+?src=|
1488 data-video-url=|
1489 <embed[^>]+?src=|
1490 embedSWF\(?:\s*|
1491 <object[^>]+data=|
1492 new\s+SWFObject\(
1493 )
1494 (["\'])
1495 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1496 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1497 \1''', webpage)]
1498
1499 # lazyYT YouTube embed
1500 entries.extend(list(map(
1501 unescapeHTML,
1502 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1503
1504 # Wordpress "YouTube Video Importer" plugin
1505 matches = re.findall(r'''(?x)<div[^>]+
1506 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1507 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1508 entries.extend(m[-1] for m in matches)
1509
1510 return entries
1511
1512 @staticmethod
1513 def _extract_url(webpage):
1514 urls = YoutubeIE._extract_urls(webpage)
1515 return urls[0] if urls else None
1516
1517 @classmethod
1518 def extract_id(cls, url):
1519 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1520 if mobj is None:
1521 raise ExtractorError('Invalid URL: %s' % url)
1522 video_id = mobj.group(2)
1523 return video_id
1524
1525 def _extract_chapters_from_json(self, data, video_id, duration):
1526 chapters_list = try_get(
1527 data,
1528 lambda x: x['playerOverlays']
1529 ['playerOverlayRenderer']
1530 ['decoratedPlayerBarRenderer']
1531 ['decoratedPlayerBarRenderer']
1532 ['playerBar']
1533 ['chapteredPlayerBarRenderer']
1534 ['chapters'],
1535 list)
1536 if not chapters_list:
1537 return
1538
1539 def chapter_time(chapter):
1540 return float_or_none(
1541 try_get(
1542 chapter,
1543 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1544 int),
1545 scale=1000)
1546 chapters = []
1547 for next_num, chapter in enumerate(chapters_list, start=1):
1548 start_time = chapter_time(chapter)
1549 if start_time is None:
1550 continue
1551 end_time = (chapter_time(chapters_list[next_num])
1552 if next_num < len(chapters_list) else duration)
1553 if end_time is None:
1554 continue
1555 title = try_get(
1556 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1557 compat_str)
1558 chapters.append({
1559 'start_time': start_time,
1560 'end_time': end_time,
1561 'title': title,
1562 })
1563 return chapters
1564
1565 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
1566 return self._parse_json(self._search_regex(
1567 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
1568 regex), webpage, name, default='{}'), video_id, fatal=False)
1569
1570 @staticmethod
1571 def parse_time_text(time_text):
1572 """
1573 Parse the comment time text
1574 time_text is in the format 'X units ago (edited)'
1575 """
1576 time_text_split = time_text.split(' ')
1577 if len(time_text_split) >= 3:
1578 return datetime_from_str('now-%s%s' % (time_text_split[0], time_text_split[1]), precision='auto')
1579
1580 @staticmethod
1581 def _join_text_entries(runs):
1582 text = None
1583 for run in runs:
1584 if not isinstance(run, dict):
1585 continue
1586 sub_text = try_get(run, lambda x: x['text'], compat_str)
1587 if sub_text:
1588 if not text:
1589 text = sub_text
1590 continue
1591 text += sub_text
1592 return text
1593
1594 def _extract_comment(self, comment_renderer, parent=None):
1595 comment_id = comment_renderer.get('commentId')
1596 if not comment_id:
1597 return
1598 comment_text_runs = try_get(comment_renderer, lambda x: x['contentText']['runs']) or []
1599 text = self._join_text_entries(comment_text_runs) or ''
1600 comment_time_text = try_get(comment_renderer, lambda x: x['publishedTimeText']['runs']) or []
1601 time_text = self._join_text_entries(comment_time_text)
1602 timestamp = calendar.timegm(self.parse_time_text(time_text).timetuple())
1603 author = try_get(comment_renderer, lambda x: x['authorText']['simpleText'], compat_str)
1604 author_id = try_get(comment_renderer,
1605 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
1606 votes = str_to_int(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
1607 lambda x: x['likeCount']), compat_str)) or 0
1608 author_thumbnail = try_get(comment_renderer,
1609 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
1610
1611 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
1612 is_liked = try_get(comment_renderer, lambda x: x['isLiked'], bool)
1613 return {
1614 'id': comment_id,
1615 'text': text,
1616 'timestamp': timestamp,
1617 'time_text': time_text,
1618 'like_count': votes,
1619 'is_favorited': is_liked,
1620 'author': author,
1621 'author_id': author_id,
1622 'author_thumbnail': author_thumbnail,
1623 'author_is_uploader': author_is_uploader,
1624 'parent': parent or 'root'
1625 }
1626
1627 def _comment_entries(self, root_continuation_data, identity_token, account_syncid,
1628 ytcfg, session_token_list, parent=None, comment_counts=None):
1629
1630 def extract_thread(parent_renderer):
1631 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
1632 if not parent:
1633 comment_counts[2] = 0
1634 for content in contents:
1635 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
1636 comment_renderer = try_get(
1637 comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
1638 content, (lambda x: x['commentRenderer'], dict))
1639
1640 if not comment_renderer:
1641 continue
1642 comment = self._extract_comment(comment_renderer, parent)
1643 if not comment:
1644 continue
1645 comment_counts[0] += 1
1646 yield comment
1647 # Attempt to get the replies
1648 comment_replies_renderer = try_get(
1649 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
1650
1651 if comment_replies_renderer:
1652 comment_counts[2] += 1
1653 comment_entries_iter = self._comment_entries(
1654 comment_replies_renderer, identity_token, account_syncid, ytcfg,
1655 parent=comment.get('id'), session_token_list=session_token_list,
1656 comment_counts=comment_counts)
1657
1658 for reply_comment in comment_entries_iter:
1659 yield reply_comment
1660
1661 if not comment_counts:
1662 # comment so far, est. total comments, current comment thread #
1663 comment_counts = [0, 0, 0]
1664
1665 # TODO: Generalize the download code with TabIE
1666 context = self._extract_context(ytcfg)
1667 visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
1668 continuation = YoutubeTabIE._extract_continuation(root_continuation_data) # TODO
1669 first_continuation = False
1670 if parent is None:
1671 first_continuation = True
1672
1673 for page_num in itertools.count(0):
1674 if not continuation:
1675 break
1676 headers = self._generate_api_headers(ytcfg, identity_token, account_syncid, visitor_data)
1677 retries = self._downloader.params.get('extractor_retries', 3)
1678 count = -1
1679 last_error = None
1680
1681 while count < retries:
1682 count += 1
1683 if last_error:
1684 self.report_warning('%s. Retrying ...' % last_error)
1685 try:
1686 query = {
1687 'ctoken': continuation['ctoken'],
1688 'pbj': 1,
1689 'type': 'next',
1690 }
1691 if parent:
1692 query['action_get_comment_replies'] = 1
1693 else:
1694 query['action_get_comments'] = 1
1695
1696 comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
1697 if page_num == 0:
1698 if first_continuation:
1699 note_prefix = 'Downloading initial comment continuation page'
1700 else:
1701 note_prefix = ' Downloading comment reply thread %d %s' % (comment_counts[2], comment_prog_str)
1702 else:
1703 note_prefix = '%sDownloading comment%s page %d %s' % (
1704 ' ' if parent else '',
1705 ' replies' if parent else '',
1706 page_num,
1707 comment_prog_str)
1708
1709 browse = self._download_json(
1710 'https://www.youtube.com/comment_service_ajax', None,
1711 '%s %s' % (note_prefix, '(retry #%d)' % count if count else ''),
1712 headers=headers, query=query,
1713 data=urlencode_postdata({
1714 'session_token': session_token_list[0]
1715 }))
1716 except ExtractorError as e:
1717 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404, 413):
1718 if e.cause.code == 413:
1719 self.report_warning('Assumed end of comments (received HTTP Error 413)')
1720 return
1721 # Downloading page may result in intermittent 5xx HTTP error
1722 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
1723 last_error = 'HTTP Error %s' % e.cause.code
1724 if e.cause.code == 404:
1725 last_error = last_error + ' (this API is probably deprecated)'
1726 if count < retries:
1727 continue
1728 raise
1729 else:
1730 session_token = try_get(browse, lambda x: x['xsrf_token'], compat_str)
1731 if session_token:
1732 session_token_list[0] = session_token
1733
1734 response = try_get(browse,
1735 (lambda x: x['response'],
1736 lambda x: x[1]['response'])) or {}
1737
1738 if response.get('continuationContents'):
1739 break
1740
1741 # YouTube sometimes gives reload: now json if something went wrong (e.g. bad auth)
1742 if browse.get('reload'):
1743 raise ExtractorError('Invalid or missing params in continuation request', expected=False)
1744
1745 # TODO: not tested, merged from old extractor
1746 err_msg = browse.get('externalErrorMessage')
1747 if err_msg:
1748 raise ExtractorError('YouTube said: %s' % err_msg, expected=False)
1749
1750 # Youtube sometimes sends incomplete data
1751 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1752 last_error = 'Incomplete data received'
1753 if count >= retries:
1754 raise ExtractorError(last_error)
1755
1756 if not response:
1757 break
1758 visitor_data = try_get(
1759 response,
1760 lambda x: x['responseContext']['webResponseContextExtensionData']['ytConfigData']['visitorData'],
1761 compat_str) or visitor_data
1762
1763 known_continuation_renderers = {
1764 'itemSectionContinuation': extract_thread,
1765 'commentRepliesContinuation': extract_thread
1766 }
1767
1768 # extract next root continuation from the results
1769 continuation_contents = try_get(
1770 response, lambda x: x['continuationContents'], dict) or {}
1771
1772 for key, value in continuation_contents.items():
1773 if key not in known_continuation_renderers:
1774 continue
1775 continuation_renderer = value
1776
1777 if first_continuation:
1778 first_continuation = False
1779 expected_comment_count = try_get(
1780 continuation_renderer,
1781 (lambda x: x['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'],
1782 lambda x: x['header']['commentsHeaderRenderer']['commentsCount']['runs'][0]['text']),
1783 compat_str)
1784
1785 if expected_comment_count:
1786 comment_counts[1] = str_to_int(expected_comment_count)
1787 self.to_screen('Downloading ~%d comments' % str_to_int(expected_comment_count))
1788 yield comment_counts[1]
1789
1790 # TODO: cli arg.
1791 # 1/True for newest, 0/False for popular (default)
1792 comment_sort_index = int(True)
1793 sort_continuation_renderer = try_get(
1794 continuation_renderer,
1795 lambda x: x['header']['commentsHeaderRenderer']['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems']
1796 [comment_sort_index]['continuation']['reloadContinuationData'], dict)
1797 # If this fails, the initial continuation page
1798 # starts off with popular anyways.
1799 if sort_continuation_renderer:
1800 continuation = YoutubeTabIE._build_continuation_query(
1801 continuation=sort_continuation_renderer.get('continuation'),
1802 ctp=sort_continuation_renderer.get('clickTrackingParams'))
1803 self.to_screen('Sorting comments by %s' % ('popular' if comment_sort_index == 0 else 'newest'))
1804 break
1805
1806 for entry in known_continuation_renderers[key](continuation_renderer):
1807 yield entry
1808
1809 continuation = YoutubeTabIE._extract_continuation(continuation_renderer) # TODO
1810 break
1811
1812 def _extract_comments(self, ytcfg, video_id, contents, webpage, xsrf_token):
1813 """Entry for comment extraction"""
1814 comments = []
1815 known_entry_comment_renderers = (
1816 'itemSectionRenderer',
1817 )
1818 estimated_total = 0
1819 for entry in contents:
1820 for key, renderer in entry.items():
1821 if key not in known_entry_comment_renderers:
1822 continue
1823
1824 comment_iter = self._comment_entries(
1825 renderer,
1826 identity_token=self._extract_identity_token(webpage, item_id=video_id),
1827 account_syncid=self._extract_account_syncid(ytcfg),
1828 ytcfg=ytcfg,
1829 session_token_list=[xsrf_token])
1830
1831 for comment in comment_iter:
1832 if isinstance(comment, int):
1833 estimated_total = comment
1834 continue
1835 comments.append(comment)
1836 break
1837 self.to_screen('Downloaded %d/%d comments' % (len(comments), estimated_total))
1838 return {
1839 'comments': comments,
1840 'comment_count': len(comments),
1841 }
1842
1843 def _real_extract(self, url):
1844 url, smuggled_data = unsmuggle_url(url, {})
1845 video_id = self._match_id(url)
1846 base_url = self.http_scheme() + '//www.youtube.com/'
1847 webpage_url = base_url + 'watch?v=' + video_id
1848 webpage = self._download_webpage(
1849 webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False)
1850
1851 player_response = None
1852 if webpage:
1853 player_response = self._extract_yt_initial_variable(
1854 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
1855 video_id, 'initial player response')
1856
1857 ytcfg = self._extract_ytcfg(video_id, webpage)
1858 if not player_response:
1859 player_response = self._call_api(
1860 'player', {'videoId': video_id}, video_id, api_key=self._extract_api_key(ytcfg))
1861
1862 playability_status = player_response.get('playabilityStatus') or {}
1863 if playability_status.get('reason') == 'Sign in to confirm your age':
1864 pr = self._parse_json(try_get(compat_parse_qs(
1865 self._download_webpage(
1866 base_url + 'get_video_info', video_id,
1867 'Refetching age-gated info webpage',
1868 'unable to download video info webpage', query={
1869 'video_id': video_id,
1870 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1871 }, fatal=False)),
1872 lambda x: x['player_response'][0],
1873 compat_str) or '{}', video_id)
1874 if pr:
1875 player_response = pr
1876
1877 trailer_video_id = try_get(
1878 playability_status,
1879 lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'],
1880 compat_str)
1881 if trailer_video_id:
1882 return self.url_result(
1883 trailer_video_id, self.ie_key(), trailer_video_id)
1884
1885 def get_text(x):
1886 if not x:
1887 return
1888 text = x.get('simpleText')
1889 if text and isinstance(text, compat_str):
1890 return text
1891 runs = x.get('runs')
1892 if not isinstance(runs, list):
1893 return
1894 return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
1895
1896 search_meta = (
1897 lambda x: self._html_search_meta(x, webpage, default=None)) \
1898 if webpage else lambda x: None
1899
1900 video_details = player_response.get('videoDetails') or {}
1901 microformat = try_get(
1902 player_response,
1903 lambda x: x['microformat']['playerMicroformatRenderer'],
1904 dict) or {}
1905 video_title = video_details.get('title') \
1906 or get_text(microformat.get('title')) \
1907 or search_meta(['og:title', 'twitter:title', 'title'])
1908 video_description = video_details.get('shortDescription')
1909
1910 if not smuggled_data.get('force_singlefeed', False):
1911 if not self._downloader.params.get('noplaylist'):
1912 multifeed_metadata_list = try_get(
1913 player_response,
1914 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1915 compat_str)
1916 if multifeed_metadata_list:
1917 entries = []
1918 feed_ids = []
1919 for feed in multifeed_metadata_list.split(','):
1920 # Unquote should take place before split on comma (,) since textual
1921 # fields may contain comma as well (see
1922 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1923 feed_data = compat_parse_qs(
1924 compat_urllib_parse_unquote_plus(feed))
1925
1926 def feed_entry(name):
1927 return try_get(
1928 feed_data, lambda x: x[name][0], compat_str)
1929
1930 feed_id = feed_entry('id')
1931 if not feed_id:
1932 continue
1933 feed_title = feed_entry('title')
1934 title = video_title
1935 if feed_title:
1936 title += ' (%s)' % feed_title
1937 entries.append({
1938 '_type': 'url_transparent',
1939 'ie_key': 'Youtube',
1940 'url': smuggle_url(
1941 base_url + 'watch?v=' + feed_data['id'][0],
1942 {'force_singlefeed': True}),
1943 'title': title,
1944 })
1945 feed_ids.append(feed_id)
1946 self.to_screen(
1947 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1948 % (', '.join(feed_ids), video_id))
1949 return self.playlist_result(
1950 entries, video_id, video_title, video_description)
1951 else:
1952 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1953
1954 formats = []
1955 itags = []
1956 itag_qualities = {}
1957 player_url = None
1958 q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'])
1959 streaming_data = player_response.get('streamingData') or {}
1960 streaming_formats = streaming_data.get('formats') or []
1961 streaming_formats.extend(streaming_data.get('adaptiveFormats') or [])
1962 for fmt in streaming_formats:
1963 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
1964 continue
1965
1966 itag = str_or_none(fmt.get('itag'))
1967 quality = fmt.get('quality')
1968 if itag and quality:
1969 itag_qualities[itag] = quality
1970 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
1971 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
1972 # number of fragment that would subsequently requested with (`&sq=N`)
1973 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
1974 continue
1975
1976 fmt_url = fmt.get('url')
1977 if not fmt_url:
1978 sc = compat_parse_qs(fmt.get('signatureCipher'))
1979 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
1980 encrypted_sig = try_get(sc, lambda x: x['s'][0])
1981 if not (sc and fmt_url and encrypted_sig):
1982 continue
1983 if not player_url:
1984 if not webpage:
1985 continue
1986 player_url = self._search_regex(
1987 r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"',
1988 webpage, 'player URL', fatal=False)
1989 if not player_url:
1990 continue
1991 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
1992 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
1993 fmt_url += '&' + sp + '=' + signature
1994
1995 if itag:
1996 itags.append(itag)
1997 tbr = float_or_none(
1998 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
1999 dct = {
2000 'asr': int_or_none(fmt.get('audioSampleRate')),
2001 'filesize': int_or_none(fmt.get('contentLength')),
2002 'format_id': itag,
2003 'format_note': fmt.get('qualityLabel') or quality,
2004 'fps': int_or_none(fmt.get('fps')),
2005 'height': int_or_none(fmt.get('height')),
2006 'quality': q(quality),
2007 'tbr': tbr,
2008 'url': fmt_url,
2009 'width': fmt.get('width'),
2010 }
2011 mimetype = fmt.get('mimeType')
2012 if mimetype:
2013 mobj = re.match(
2014 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype)
2015 if mobj:
2016 dct['ext'] = mimetype2ext(mobj.group(1))
2017 dct.update(parse_codecs(mobj.group(2)))
2018 no_audio = dct.get('acodec') == 'none'
2019 no_video = dct.get('vcodec') == 'none'
2020 if no_audio:
2021 dct['vbr'] = tbr
2022 if no_video:
2023 dct['abr'] = tbr
2024 if no_audio or no_video:
2025 dct['downloader_options'] = {
2026 # Youtube throttles chunks >~10M
2027 'http_chunk_size': 10485760,
2028 }
2029 if dct.get('ext'):
2030 dct['container'] = dct['ext'] + '_dash'
2031 formats.append(dct)
2032
2033 hls_manifest_url = streaming_data.get('hlsManifestUrl')
2034 if hls_manifest_url:
2035 for f in self._extract_m3u8_formats(
2036 hls_manifest_url, video_id, 'mp4', fatal=False):
2037 itag = self._search_regex(
2038 r'/itag/(\d+)', f['url'], 'itag', default=None)
2039 if itag:
2040 f['format_id'] = itag
2041 formats.append(f)
2042
2043 if self._downloader.params.get('youtube_include_dash_manifest', True):
2044 dash_manifest_url = streaming_data.get('dashManifestUrl')
2045 if dash_manifest_url:
2046 for f in self._extract_mpd_formats(
2047 dash_manifest_url, video_id, fatal=False):
2048 itag = f['format_id']
2049 if itag in itags:
2050 continue
2051 if itag in itag_qualities:
2052 # Not actually usefull since the sorting is already done with "quality,res,fps,codec"
2053 # but kept to maintain feature parity (and code similarity) with youtube-dl
2054 # Remove if this causes any issues with sorting in future
2055 f['quality'] = q(itag_qualities[itag])
2056 filesize = int_or_none(self._search_regex(
2057 r'/clen/(\d+)', f.get('fragment_base_url')
2058 or f['url'], 'file size', default=None))
2059 if filesize:
2060 f['filesize'] = filesize
2061 formats.append(f)
2062
2063 if not formats:
2064 if not self._downloader.params.get('allow_unplayable_formats') and streaming_data.get('licenseInfos'):
2065 self.raise_no_formats(
2066 'This video is DRM protected.', expected=True)
2067 pemr = try_get(
2068 playability_status,
2069 lambda x: x['errorScreen']['playerErrorMessageRenderer'],
2070 dict) or {}
2071 reason = get_text(pemr.get('reason')) or playability_status.get('reason')
2072 subreason = pemr.get('subreason')
2073 if subreason:
2074 subreason = clean_html(get_text(subreason))
2075 if subreason == 'The uploader has not made this video available in your country.':
2076 countries = microformat.get('availableCountries')
2077 if not countries:
2078 regions_allowed = search_meta('regionsAllowed')
2079 countries = regions_allowed.split(',') if regions_allowed else None
2080 self.raise_geo_restricted(subreason, countries, metadata_available=True)
2081 reason += '\n' + subreason
2082 if reason:
2083 self.raise_no_formats(reason, expected=True)
2084
2085 self._sort_formats(formats)
2086
2087 keywords = video_details.get('keywords') or []
2088 if not keywords and webpage:
2089 keywords = [
2090 unescapeHTML(m.group('content'))
2091 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
2092 for keyword in keywords:
2093 if keyword.startswith('yt:stretch='):
2094 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
2095 if mobj:
2096 # NB: float is intentional for forcing float division
2097 w, h = (float(v) for v in mobj.groups())
2098 if w > 0 and h > 0:
2099 ratio = w / h
2100 for f in formats:
2101 if f.get('vcodec') != 'none':
2102 f['stretched_ratio'] = ratio
2103 break
2104
2105 thumbnails = []
2106 for container in (video_details, microformat):
2107 for thumbnail in (try_get(
2108 container,
2109 lambda x: x['thumbnail']['thumbnails'], list) or []):
2110 thumbnail_url = thumbnail.get('url')
2111 if not thumbnail_url:
2112 continue
2113 # Sometimes youtube gives a wrong thumbnail URL. See:
2114 # https://github.com/yt-dlp/yt-dlp/issues/233
2115 # https://github.com/ytdl-org/youtube-dl/issues/28023
2116 if 'maxresdefault' in thumbnail_url:
2117 thumbnail_url = thumbnail_url.split('?')[0]
2118 thumbnails.append({
2119 'height': int_or_none(thumbnail.get('height')),
2120 'url': thumbnail_url,
2121 'width': int_or_none(thumbnail.get('width')),
2122 })
2123 if thumbnails:
2124 break
2125 else:
2126 thumbnail = search_meta(['og:image', 'twitter:image'])
2127 if thumbnail:
2128 thumbnails = [{'url': thumbnail}]
2129
2130 category = microformat.get('category') or search_meta('genre')
2131 channel_id = video_details.get('channelId') \
2132 or microformat.get('externalChannelId') \
2133 or search_meta('channelId')
2134 duration = int_or_none(
2135 video_details.get('lengthSeconds')
2136 or microformat.get('lengthSeconds')) \
2137 or parse_duration(search_meta('duration'))
2138 is_live = video_details.get('isLive')
2139 owner_profile_url = microformat.get('ownerProfileUrl')
2140
2141 info = {
2142 'id': video_id,
2143 'title': self._live_title(video_title) if is_live else video_title,
2144 'formats': formats,
2145 'thumbnails': thumbnails,
2146 'description': video_description,
2147 'upload_date': unified_strdate(
2148 microformat.get('uploadDate')
2149 or search_meta('uploadDate')),
2150 'uploader': video_details['author'],
2151 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
2152 'uploader_url': owner_profile_url,
2153 'channel_id': channel_id,
2154 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None,
2155 'duration': duration,
2156 'view_count': int_or_none(
2157 video_details.get('viewCount')
2158 or microformat.get('viewCount')
2159 or search_meta('interactionCount')),
2160 'average_rating': float_or_none(video_details.get('averageRating')),
2161 'age_limit': 18 if (
2162 microformat.get('isFamilySafe') is False
2163 or search_meta('isFamilyFriendly') == 'false'
2164 or search_meta('og:restrictions:age') == '18+') else 0,
2165 'webpage_url': webpage_url,
2166 'categories': [category] if category else None,
2167 'tags': keywords,
2168 'is_live': is_live,
2169 'playable_in_embed': playability_status.get('playableInEmbed'),
2170 'was_live': video_details.get('isLiveContent'),
2171 }
2172
2173 pctr = try_get(
2174 player_response,
2175 lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
2176 subtitles = {}
2177 if pctr:
2178 def process_language(container, base_url, lang_code, query):
2179 lang_subs = []
2180 for fmt in self._SUBTITLE_FORMATS:
2181 query.update({
2182 'fmt': fmt,
2183 })
2184 lang_subs.append({
2185 'ext': fmt,
2186 'url': update_url_query(base_url, query),
2187 })
2188 container[lang_code] = lang_subs
2189
2190 for caption_track in (pctr.get('captionTracks') or []):
2191 base_url = caption_track.get('baseUrl')
2192 if not base_url:
2193 continue
2194 if caption_track.get('kind') != 'asr':
2195 lang_code = caption_track.get('languageCode')
2196 if not lang_code:
2197 continue
2198 process_language(
2199 subtitles, base_url, lang_code, {})
2200 continue
2201 automatic_captions = {}
2202 for translation_language in (pctr.get('translationLanguages') or []):
2203 translation_language_code = translation_language.get('languageCode')
2204 if not translation_language_code:
2205 continue
2206 process_language(
2207 automatic_captions, base_url, translation_language_code,
2208 {'tlang': translation_language_code})
2209 info['automatic_captions'] = automatic_captions
2210 info['subtitles'] = subtitles
2211
2212 parsed_url = compat_urllib_parse_urlparse(url)
2213 for component in [parsed_url.fragment, parsed_url.query]:
2214 query = compat_parse_qs(component)
2215 for k, v in query.items():
2216 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
2217 d_k += '_time'
2218 if d_k not in info and k in s_ks:
2219 info[d_k] = parse_duration(query[k][0])
2220
2221 # Youtube Music Auto-generated description
2222 if video_description:
2223 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)
2224 if mobj:
2225 release_year = mobj.group('release_year')
2226 release_date = mobj.group('release_date')
2227 if release_date:
2228 release_date = release_date.replace('-', '')
2229 if not release_year:
2230 release_year = release_date[:4]
2231 info.update({
2232 'album': mobj.group('album'.strip()),
2233 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
2234 'track': mobj.group('track').strip(),
2235 'release_date': release_date,
2236 'release_year': int_or_none(release_year),
2237 })
2238
2239 initial_data = None
2240 if webpage:
2241 initial_data = self._extract_yt_initial_variable(
2242 webpage, self._YT_INITIAL_DATA_RE, video_id,
2243 'yt initial data')
2244 if not initial_data:
2245 initial_data = self._call_api(
2246 'next', {'videoId': video_id}, video_id, fatal=False, api_key=self._extract_api_key(ytcfg))
2247
2248 if not is_live:
2249 try:
2250 # This will error if there is no livechat
2251 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2252 info['subtitles']['live_chat'] = [{
2253 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
2254 'video_id': video_id,
2255 'ext': 'json',
2256 'protocol': 'youtube_live_chat_replay',
2257 }]
2258 except (KeyError, IndexError, TypeError):
2259 pass
2260
2261 if initial_data:
2262 chapters = self._extract_chapters_from_json(
2263 initial_data, video_id, duration)
2264 if not chapters:
2265 for engagment_pannel in (initial_data.get('engagementPanels') or []):
2266 contents = try_get(
2267 engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'],
2268 list)
2269 if not contents:
2270 continue
2271
2272 def chapter_time(mmlir):
2273 return parse_duration(
2274 get_text(mmlir.get('timeDescription')))
2275
2276 chapters = []
2277 for next_num, content in enumerate(contents, start=1):
2278 mmlir = content.get('macroMarkersListItemRenderer') or {}
2279 start_time = chapter_time(mmlir)
2280 end_time = chapter_time(try_get(
2281 contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \
2282 if next_num < len(contents) else duration
2283 if start_time is None or end_time is None:
2284 continue
2285 chapters.append({
2286 'start_time': start_time,
2287 'end_time': end_time,
2288 'title': get_text(mmlir.get('title')),
2289 })
2290 if chapters:
2291 break
2292 if chapters:
2293 info['chapters'] = chapters
2294
2295 contents = try_get(
2296 initial_data,
2297 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
2298 list) or []
2299 for content in contents:
2300 vpir = content.get('videoPrimaryInfoRenderer')
2301 if vpir:
2302 stl = vpir.get('superTitleLink')
2303 if stl:
2304 stl = get_text(stl)
2305 if try_get(
2306 vpir,
2307 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
2308 info['location'] = stl
2309 else:
2310 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
2311 if mobj:
2312 info.update({
2313 'series': mobj.group(1),
2314 'season_number': int(mobj.group(2)),
2315 'episode_number': int(mobj.group(3)),
2316 })
2317 for tlb in (try_get(
2318 vpir,
2319 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
2320 list) or []):
2321 tbr = tlb.get('toggleButtonRenderer') or {}
2322 for getter, regex in [(
2323 lambda x: x['defaultText']['accessibility']['accessibilityData'],
2324 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
2325 lambda x: x['accessibility'],
2326 lambda x: x['accessibilityData']['accessibilityData'],
2327 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
2328 label = (try_get(tbr, getter, dict) or {}).get('label')
2329 if label:
2330 mobj = re.match(regex, label)
2331 if mobj:
2332 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
2333 break
2334 sbr_tooltip = try_get(
2335 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
2336 if sbr_tooltip:
2337 like_count, dislike_count = sbr_tooltip.split(' / ')
2338 info.update({
2339 'like_count': str_to_int(like_count),
2340 'dislike_count': str_to_int(dislike_count),
2341 })
2342 vsir = content.get('videoSecondaryInfoRenderer')
2343 if vsir:
2344 info['channel'] = get_text(try_get(
2345 vsir,
2346 lambda x: x['owner']['videoOwnerRenderer']['title'],
2347 dict))
2348 rows = try_get(
2349 vsir,
2350 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
2351 list) or []
2352 multiple_songs = False
2353 for row in rows:
2354 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2355 multiple_songs = True
2356 break
2357 for row in rows:
2358 mrr = row.get('metadataRowRenderer') or {}
2359 mrr_title = mrr.get('title')
2360 if not mrr_title:
2361 continue
2362 mrr_title = get_text(mrr['title'])
2363 mrr_contents_text = get_text(mrr['contents'][0])
2364 if mrr_title == 'License':
2365 info['license'] = mrr_contents_text
2366 elif not multiple_songs:
2367 if mrr_title == 'Album':
2368 info['album'] = mrr_contents_text
2369 elif mrr_title == 'Artist':
2370 info['artist'] = mrr_contents_text
2371 elif mrr_title == 'Song':
2372 info['track'] = mrr_contents_text
2373
2374 fallbacks = {
2375 'channel': 'uploader',
2376 'channel_id': 'uploader_id',
2377 'channel_url': 'uploader_url',
2378 }
2379 for to, frm in fallbacks.items():
2380 if not info.get(to):
2381 info[to] = info.get(frm)
2382
2383 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
2384 v = info.get(s_k)
2385 if v:
2386 info[d_k] = v
2387
2388 is_private = bool_or_none(video_details.get('isPrivate'))
2389 is_unlisted = bool_or_none(microformat.get('isUnlisted'))
2390 is_membersonly = None
2391 is_premium = None
2392 if initial_data and is_private is not None:
2393 is_membersonly = False
2394 is_premium = False
2395 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list)
2396 for content in contents or []:
2397 badges = try_get(content, lambda x: x['videoPrimaryInfoRenderer']['badges'], list)
2398 for badge in badges or []:
2399 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label']) or ''
2400 if label.lower() == 'members only':
2401 is_membersonly = True
2402 break
2403 elif label.lower() == 'premium':
2404 is_premium = True
2405 break
2406 if is_membersonly or is_premium:
2407 break
2408
2409 # TODO: Add this for playlists
2410 info['availability'] = self._availability(
2411 is_private=is_private,
2412 needs_premium=is_premium,
2413 needs_subscription=is_membersonly,
2414 needs_auth=info['age_limit'] >= 18,
2415 is_unlisted=None if is_private is None else is_unlisted)
2416
2417 # get xsrf for annotations or comments
2418 get_annotations = self._downloader.params.get('writeannotations', False)
2419 get_comments = self._downloader.params.get('getcomments', False)
2420 if get_annotations or get_comments:
2421 xsrf_token = None
2422 ytcfg = self._extract_ytcfg(video_id, webpage)
2423 if ytcfg:
2424 xsrf_token = try_get(ytcfg, lambda x: x['XSRF_TOKEN'], compat_str)
2425 if not xsrf_token:
2426 xsrf_token = self._search_regex(
2427 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>(?:(?!\2).)+)\2',
2428 webpage, 'xsrf token', group='xsrf_token', fatal=False)
2429
2430 # annotations
2431 if get_annotations:
2432 invideo_url = try_get(
2433 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2434 if xsrf_token and invideo_url:
2435 xsrf_field_name = None
2436 if ytcfg:
2437 xsrf_field_name = try_get(ytcfg, lambda x: x['XSRF_FIELD_NAME'], compat_str)
2438 if not xsrf_field_name:
2439 xsrf_field_name = self._search_regex(
2440 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2441 webpage, 'xsrf field name',
2442 group='xsrf_field_name', default='session_token')
2443 info['annotations'] = self._download_webpage(
2444 self._proto_relative_url(invideo_url),
2445 video_id, note='Downloading annotations',
2446 errnote='Unable to download video annotations', fatal=False,
2447 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2448
2449 if get_comments:
2450 info['__post_extractor'] = lambda: self._extract_comments(ytcfg, video_id, contents, webpage, xsrf_token)
2451
2452 self.mark_watched(video_id, player_response)
2453
2454 return info
2455
2456
2457 class YoutubeTabIE(YoutubeBaseInfoExtractor):
2458 IE_DESC = 'YouTube.com tab'
2459 _VALID_URL = r'''(?x)
2460 https?://
2461 (?:\w+\.)?
2462 (?:
2463 youtube(?:kids)?\.com|
2464 invidio\.us
2465 )/
2466 (?:
2467 (?:channel|c|user)/|
2468 (?P<not_channel>
2469 feed/|hashtag/|
2470 (?:playlist|watch)\?.*?\blist=
2471 )|
2472 (?!(?:%s)\b) # Direct URLs
2473 )
2474 (?P<id>[^/?\#&]+)
2475 ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
2476 IE_NAME = 'youtube:tab'
2477
2478 _TESTS = [{
2479 # playlists, multipage
2480 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
2481 'playlist_mincount': 94,
2482 'info_dict': {
2483 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2484 'title': 'Игорь Клейнер - Playlists',
2485 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2486 'uploader': 'Игорь Клейнер',
2487 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2488 },
2489 }, {
2490 # playlists, multipage, different order
2491 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2492 'playlist_mincount': 94,
2493 'info_dict': {
2494 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2495 'title': 'Игорь Клейнер - Playlists',
2496 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2497 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2498 'uploader': 'Игорь Клейнер',
2499 },
2500 }, {
2501 # playlists, series
2502 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
2503 'playlist_mincount': 5,
2504 'info_dict': {
2505 'id': 'UCYO_jab_esuFRV4b17AJtAw',
2506 'title': '3Blue1Brown - Playlists',
2507 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
2508 },
2509 }, {
2510 # playlists, singlepage
2511 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2512 'playlist_mincount': 4,
2513 'info_dict': {
2514 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2515 'title': 'ThirstForScience - Playlists',
2516 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
2517 'uploader': 'ThirstForScience',
2518 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2519 }
2520 }, {
2521 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
2522 'only_matching': True,
2523 }, {
2524 # basic, single video playlist
2525 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2526 'info_dict': {
2527 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2528 'uploader': 'Sergey M.',
2529 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2530 'title': 'youtube-dl public playlist',
2531 },
2532 'playlist_count': 1,
2533 }, {
2534 # empty playlist
2535 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2536 'info_dict': {
2537 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2538 'uploader': 'Sergey M.',
2539 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2540 'title': 'youtube-dl empty playlist',
2541 },
2542 'playlist_count': 0,
2543 }, {
2544 # Home tab
2545 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
2546 'info_dict': {
2547 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2548 'title': 'lex will - Home',
2549 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2550 'uploader': 'lex will',
2551 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2552 },
2553 'playlist_mincount': 2,
2554 }, {
2555 # Videos tab
2556 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
2557 'info_dict': {
2558 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2559 'title': 'lex will - Videos',
2560 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2561 'uploader': 'lex will',
2562 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2563 },
2564 'playlist_mincount': 975,
2565 }, {
2566 # Videos tab, sorted by popular
2567 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
2568 'info_dict': {
2569 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2570 'title': 'lex will - Videos',
2571 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2572 'uploader': 'lex will',
2573 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2574 },
2575 'playlist_mincount': 199,
2576 }, {
2577 # Playlists tab
2578 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
2579 'info_dict': {
2580 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2581 'title': 'lex will - Playlists',
2582 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2583 'uploader': 'lex will',
2584 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2585 },
2586 'playlist_mincount': 17,
2587 }, {
2588 # Community tab
2589 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
2590 'info_dict': {
2591 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2592 'title': 'lex will - Community',
2593 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2594 'uploader': 'lex will',
2595 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2596 },
2597 'playlist_mincount': 18,
2598 }, {
2599 # Channels tab
2600 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
2601 'info_dict': {
2602 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2603 'title': 'lex will - Channels',
2604 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2605 'uploader': 'lex will',
2606 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2607 },
2608 'playlist_mincount': 12,
2609 }, {
2610 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2611 'only_matching': True,
2612 }, {
2613 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2614 'only_matching': True,
2615 }, {
2616 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2617 'only_matching': True,
2618 }, {
2619 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2620 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2621 'info_dict': {
2622 'title': '29C3: Not my department',
2623 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2624 'uploader': 'Christiaan008',
2625 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
2626 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
2627 },
2628 'playlist_count': 96,
2629 }, {
2630 'note': 'Large playlist',
2631 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2632 'info_dict': {
2633 'title': 'Uploads from Cauchemar',
2634 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2635 'uploader': 'Cauchemar',
2636 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
2637 },
2638 'playlist_mincount': 1123,
2639 }, {
2640 # even larger playlist, 8832 videos
2641 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
2642 'only_matching': True,
2643 }, {
2644 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2645 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2646 'info_dict': {
2647 'title': 'Uploads from Interstellar Movie',
2648 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2649 'uploader': 'Interstellar Movie',
2650 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
2651 },
2652 'playlist_mincount': 21,
2653 }, {
2654 'note': 'Playlist with "show unavailable videos" button',
2655 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
2656 'info_dict': {
2657 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
2658 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
2659 'uploader': 'Phim Siêu Nhân Nhật Bản',
2660 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
2661 },
2662 'playlist_mincount': 1400,
2663 'expected_warnings': [
2664 'YouTube said: INFO - Unavailable videos are hidden',
2665 ]
2666 }, {
2667 'note': 'Playlist with unavailable videos in a later page',
2668 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
2669 'info_dict': {
2670 'title': 'Uploads from BlankTV',
2671 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
2672 'uploader': 'BlankTV',
2673 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
2674 },
2675 'playlist_mincount': 20000,
2676 }, {
2677 # https://github.com/ytdl-org/youtube-dl/issues/21844
2678 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2679 'info_dict': {
2680 'title': 'Data Analysis with Dr Mike Pound',
2681 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2682 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
2683 'uploader': 'Computerphile',
2684 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
2685 },
2686 'playlist_mincount': 11,
2687 }, {
2688 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2689 'only_matching': True,
2690 }, {
2691 # Playlist URL that does not actually serve a playlist
2692 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2693 'info_dict': {
2694 'id': 'FqZTN594JQw',
2695 'ext': 'webm',
2696 'title': "Smiley's People 01 detective, Adventure Series, Action",
2697 'uploader': 'STREEM',
2698 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2699 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2700 'upload_date': '20150526',
2701 'license': 'Standard YouTube License',
2702 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2703 'categories': ['People & Blogs'],
2704 'tags': list,
2705 'view_count': int,
2706 'like_count': int,
2707 'dislike_count': int,
2708 },
2709 'params': {
2710 'skip_download': True,
2711 },
2712 'skip': 'This video is not available.',
2713 'add_ie': [YoutubeIE.ie_key()],
2714 }, {
2715 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2716 'only_matching': True,
2717 }, {
2718 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
2719 'only_matching': True,
2720 }, {
2721 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
2722 'info_dict': {
2723 'id': '9Auq9mYxFEE',
2724 'ext': 'mp4',
2725 'title': compat_str,
2726 'uploader': 'Sky News',
2727 'uploader_id': 'skynews',
2728 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
2729 'upload_date': '20191102',
2730 'description': 'md5:85ddd75d888674631aaf9599a9a0b0ae',
2731 'categories': ['News & Politics'],
2732 'tags': list,
2733 'like_count': int,
2734 'dislike_count': int,
2735 },
2736 'params': {
2737 'skip_download': True,
2738 },
2739 }, {
2740 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2741 'info_dict': {
2742 'id': 'a48o2S1cPoo',
2743 'ext': 'mp4',
2744 'title': 'The Young Turks - Live Main Show',
2745 'uploader': 'The Young Turks',
2746 'uploader_id': 'TheYoungTurks',
2747 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2748 'upload_date': '20150715',
2749 'license': 'Standard YouTube License',
2750 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2751 'categories': ['News & Politics'],
2752 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2753 'like_count': int,
2754 'dislike_count': int,
2755 },
2756 'params': {
2757 'skip_download': True,
2758 },
2759 'only_matching': True,
2760 }, {
2761 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2762 'only_matching': True,
2763 }, {
2764 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2765 'only_matching': True,
2766 }, {
2767 'url': 'https://www.youtube.com/feed/trending',
2768 'only_matching': True,
2769 }, {
2770 # needs auth
2771 'url': 'https://www.youtube.com/feed/library',
2772 'only_matching': True,
2773 }, {
2774 # needs auth
2775 'url': 'https://www.youtube.com/feed/history',
2776 'only_matching': True,
2777 }, {
2778 # needs auth
2779 'url': 'https://www.youtube.com/feed/subscriptions',
2780 'only_matching': True,
2781 }, {
2782 # needs auth
2783 'url': 'https://www.youtube.com/feed/watch_later',
2784 'only_matching': True,
2785 }, {
2786 # no longer available?
2787 'url': 'https://www.youtube.com/feed/recommended',
2788 'only_matching': True,
2789 }, {
2790 # inline playlist with not always working continuations
2791 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
2792 'only_matching': True,
2793 }, {
2794 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8',
2795 'only_matching': True,
2796 }, {
2797 'url': 'https://www.youtube.com/course',
2798 'only_matching': True,
2799 }, {
2800 'url': 'https://www.youtube.com/zsecurity',
2801 'only_matching': True,
2802 }, {
2803 'url': 'http://www.youtube.com/NASAgovVideo/videos',
2804 'only_matching': True,
2805 }, {
2806 'url': 'https://www.youtube.com/TheYoungTurks/live',
2807 'only_matching': True,
2808 }, {
2809 'url': 'https://www.youtube.com/hashtag/cctv9',
2810 'info_dict': {
2811 'id': 'cctv9',
2812 'title': '#cctv9',
2813 },
2814 'playlist_mincount': 350,
2815 }, {
2816 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
2817 'only_matching': True,
2818 }]
2819
2820 @classmethod
2821 def suitable(cls, url):
2822 return False if YoutubeIE.suitable(url) else super(
2823 YoutubeTabIE, cls).suitable(url)
2824
2825 def _extract_channel_id(self, webpage):
2826 channel_id = self._html_search_meta(
2827 'channelId', webpage, 'channel id', default=None)
2828 if channel_id:
2829 return channel_id
2830 channel_url = self._html_search_meta(
2831 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
2832 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
2833 'twitter:app:url:googleplay'), webpage, 'channel url')
2834 return self._search_regex(
2835 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
2836 channel_url, 'channel id')
2837
2838 @staticmethod
2839 def _extract_basic_item_renderer(item):
2840 # Modified from _extract_grid_item_renderer
2841 known_basic_renderers = (
2842 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer'
2843 )
2844 for key, renderer in item.items():
2845 if not isinstance(renderer, dict):
2846 continue
2847 elif key in known_basic_renderers:
2848 return renderer
2849 elif key.startswith('grid') and key.endswith('Renderer'):
2850 return renderer
2851
2852 def _grid_entries(self, grid_renderer):
2853 for item in grid_renderer['items']:
2854 if not isinstance(item, dict):
2855 continue
2856 renderer = self._extract_basic_item_renderer(item)
2857 if not isinstance(renderer, dict):
2858 continue
2859 title = try_get(
2860 renderer, (lambda x: x['title']['runs'][0]['text'],
2861 lambda x: x['title']['simpleText']), compat_str)
2862 # playlist
2863 playlist_id = renderer.get('playlistId')
2864 if playlist_id:
2865 yield self.url_result(
2866 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2867 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
2868 video_title=title)
2869 continue
2870 # video
2871 video_id = renderer.get('videoId')
2872 if video_id:
2873 yield self._extract_video(renderer)
2874 continue
2875 # channel
2876 channel_id = renderer.get('channelId')
2877 if channel_id:
2878 title = try_get(
2879 renderer, lambda x: x['title']['simpleText'], compat_str)
2880 yield self.url_result(
2881 'https://www.youtube.com/channel/%s' % channel_id,
2882 ie=YoutubeTabIE.ie_key(), video_title=title)
2883 continue
2884 # generic endpoint URL support
2885 ep_url = urljoin('https://www.youtube.com/', try_get(
2886 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
2887 compat_str))
2888 if ep_url:
2889 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
2890 if ie.suitable(ep_url):
2891 yield self.url_result(
2892 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
2893 break
2894
2895 def _shelf_entries_from_content(self, shelf_renderer):
2896 content = shelf_renderer.get('content')
2897 if not isinstance(content, dict):
2898 return
2899 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
2900 if renderer:
2901 # TODO: add support for nested playlists so each shelf is processed
2902 # as separate playlist
2903 # TODO: this includes only first N items
2904 for entry in self._grid_entries(renderer):
2905 yield entry
2906 renderer = content.get('horizontalListRenderer')
2907 if renderer:
2908 # TODO
2909 pass
2910
2911 def _shelf_entries(self, shelf_renderer, skip_channels=False):
2912 ep = try_get(
2913 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
2914 compat_str)
2915 shelf_url = urljoin('https://www.youtube.com', ep)
2916 if shelf_url:
2917 # Skipping links to another channels, note that checking for
2918 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
2919 # will not work
2920 if skip_channels and '/channels?' in shelf_url:
2921 return
2922 title = try_get(
2923 shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2924 yield self.url_result(shelf_url, video_title=title)
2925 # Shelf may not contain shelf URL, fallback to extraction from content
2926 for entry in self._shelf_entries_from_content(shelf_renderer):
2927 yield entry
2928
2929 def _playlist_entries(self, video_list_renderer):
2930 for content in video_list_renderer['contents']:
2931 if not isinstance(content, dict):
2932 continue
2933 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
2934 if not isinstance(renderer, dict):
2935 continue
2936 video_id = renderer.get('videoId')
2937 if not video_id:
2938 continue
2939 yield self._extract_video(renderer)
2940
2941 def _rich_entries(self, rich_grid_renderer):
2942 renderer = try_get(
2943 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
2944 video_id = renderer.get('videoId')
2945 if not video_id:
2946 return
2947 yield self._extract_video(renderer)
2948
2949 def _video_entry(self, video_renderer):
2950 video_id = video_renderer.get('videoId')
2951 if video_id:
2952 return self._extract_video(video_renderer)
2953
2954 def _post_thread_entries(self, post_thread_renderer):
2955 post_renderer = try_get(
2956 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
2957 if not post_renderer:
2958 return
2959 # video attachment
2960 video_renderer = try_get(
2961 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
2962 video_id = video_renderer.get('videoId')
2963 if video_id:
2964 entry = self._extract_video(video_renderer)
2965 if entry:
2966 yield entry
2967 # playlist attachment
2968 playlist_id = try_get(
2969 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
2970 if playlist_id:
2971 yield self.url_result(
2972 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2973 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
2974 # inline video links
2975 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
2976 for run in runs:
2977 if not isinstance(run, dict):
2978 continue
2979 ep_url = try_get(
2980 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
2981 if not ep_url:
2982 continue
2983 if not YoutubeIE.suitable(ep_url):
2984 continue
2985 ep_video_id = YoutubeIE._match_id(ep_url)
2986 if video_id == ep_video_id:
2987 continue
2988 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
2989
2990 def _post_thread_continuation_entries(self, post_thread_continuation):
2991 contents = post_thread_continuation.get('contents')
2992 if not isinstance(contents, list):
2993 return
2994 for content in contents:
2995 renderer = content.get('backstagePostThreadRenderer')
2996 if not isinstance(renderer, dict):
2997 continue
2998 for entry in self._post_thread_entries(renderer):
2999 yield entry
3000
3001 r''' # unused
3002 def _rich_grid_entries(self, contents):
3003 for content in contents:
3004 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
3005 if video_renderer:
3006 entry = self._video_entry(video_renderer)
3007 if entry:
3008 yield entry
3009 '''
3010
3011 @staticmethod
3012 def _build_continuation_query(continuation, ctp=None):
3013 query = {
3014 'ctoken': continuation,
3015 'continuation': continuation,
3016 }
3017 if ctp:
3018 query['itct'] = ctp
3019 return query
3020
3021 @staticmethod
3022 def _extract_next_continuation_data(renderer):
3023 next_continuation = try_get(
3024 renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict)
3025 if not next_continuation:
3026 return
3027 continuation = next_continuation.get('continuation')
3028 if not continuation:
3029 return
3030 ctp = next_continuation.get('clickTrackingParams')
3031 return YoutubeTabIE._build_continuation_query(continuation, ctp)
3032
3033 @classmethod
3034 def _extract_continuation(cls, renderer):
3035 next_continuation = cls._extract_next_continuation_data(renderer)
3036 if next_continuation:
3037 return next_continuation
3038 contents = []
3039 for key in ('contents', 'items'):
3040 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
3041 for content in contents:
3042 if not isinstance(content, dict):
3043 continue
3044 continuation_ep = try_get(
3045 content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
3046 dict)
3047 if not continuation_ep:
3048 continue
3049 continuation = try_get(
3050 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
3051 if not continuation:
3052 continue
3053 ctp = continuation_ep.get('clickTrackingParams')
3054 return YoutubeTabIE._build_continuation_query(continuation, ctp)
3055
3056 def _entries(self, tab, item_id, identity_token, account_syncid, ytcfg):
3057
3058 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
3059 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
3060 for content in contents:
3061 if not isinstance(content, dict):
3062 continue
3063 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
3064 if not is_renderer:
3065 renderer = content.get('richItemRenderer')
3066 if renderer:
3067 for entry in self._rich_entries(renderer):
3068 yield entry
3069 continuation_list[0] = self._extract_continuation(parent_renderer)
3070 continue
3071 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
3072 for isr_content in isr_contents:
3073 if not isinstance(isr_content, dict):
3074 continue
3075
3076 known_renderers = {
3077 'playlistVideoListRenderer': self._playlist_entries,
3078 'gridRenderer': self._grid_entries,
3079 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'),
3080 'backstagePostThreadRenderer': self._post_thread_entries,
3081 'videoRenderer': lambda x: [self._video_entry(x)],
3082 }
3083 for key, renderer in isr_content.items():
3084 if key not in known_renderers:
3085 continue
3086 for entry in known_renderers[key](renderer):
3087 if entry:
3088 yield entry
3089 continuation_list[0] = self._extract_continuation(renderer)
3090 break
3091
3092 if not continuation_list[0]:
3093 continuation_list[0] = self._extract_continuation(is_renderer)
3094
3095 if not continuation_list[0]:
3096 continuation_list[0] = self._extract_continuation(parent_renderer)
3097
3098 continuation_list = [None] # Python 2 doesnot support nonlocal
3099 tab_content = try_get(tab, lambda x: x['content'], dict)
3100 if not tab_content:
3101 return
3102 parent_renderer = (
3103 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
3104 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
3105 for entry in extract_entries(parent_renderer):
3106 yield entry
3107 continuation = continuation_list[0]
3108 context = self._extract_context(ytcfg)
3109 visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
3110
3111 for page_num in itertools.count(1):
3112 if not continuation:
3113 break
3114 query = {
3115 'continuation': continuation['continuation'],
3116 'clickTracking': {'clickTrackingParams': continuation['itct']}
3117 }
3118 headers = self._generate_api_headers(ytcfg, identity_token, account_syncid, visitor_data)
3119 response = self._extract_response(
3120 item_id='%s page %s' % (item_id, page_num),
3121 query=query, headers=headers, ytcfg=ytcfg,
3122 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3123
3124 if not response:
3125 break
3126 visitor_data = try_get(
3127 response, lambda x: x['responseContext']['visitorData'], compat_str) or visitor_data
3128
3129 known_continuation_renderers = {
3130 'playlistVideoListContinuation': self._playlist_entries,
3131 'gridContinuation': self._grid_entries,
3132 'itemSectionContinuation': self._post_thread_continuation_entries,
3133 'sectionListContinuation': extract_entries, # for feeds
3134 }
3135 continuation_contents = try_get(
3136 response, lambda x: x['continuationContents'], dict) or {}
3137 continuation_renderer = None
3138 for key, value in continuation_contents.items():
3139 if key not in known_continuation_renderers:
3140 continue
3141 continuation_renderer = value
3142 continuation_list = [None]
3143 for entry in known_continuation_renderers[key](continuation_renderer):
3144 yield entry
3145 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
3146 break
3147 if continuation_renderer:
3148 continue
3149
3150 known_renderers = {
3151 'gridPlaylistRenderer': (self._grid_entries, 'items'),
3152 'gridVideoRenderer': (self._grid_entries, 'items'),
3153 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
3154 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
3155 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
3156 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
3157 }
3158 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3159 continuation_items = try_get(
3160 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
3161 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
3162 video_items_renderer = None
3163 for key, value in continuation_item.items():
3164 if key not in known_renderers:
3165 continue
3166 video_items_renderer = {known_renderers[key][1]: continuation_items}
3167 continuation_list = [None]
3168 for entry in known_renderers[key][0](video_items_renderer):
3169 yield entry
3170 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3171 break
3172 if video_items_renderer:
3173 continue
3174 break
3175
3176 @staticmethod
3177 def _extract_selected_tab(tabs):
3178 for tab in tabs:
3179 if try_get(tab, lambda x: x['tabRenderer']['selected'], bool):
3180 return tab['tabRenderer']
3181 else:
3182 raise ExtractorError('Unable to find selected tab')
3183
3184 @staticmethod
3185 def _extract_uploader(data):
3186 uploader = {}
3187 sidebar_renderer = try_get(
3188 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3189 if sidebar_renderer:
3190 for item in sidebar_renderer:
3191 if not isinstance(item, dict):
3192 continue
3193 renderer = item.get('playlistSidebarSecondaryInfoRenderer')
3194 if not isinstance(renderer, dict):
3195 continue
3196 owner = try_get(
3197 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3198 if owner:
3199 uploader['uploader'] = owner.get('text')
3200 uploader['uploader_id'] = try_get(
3201 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3202 uploader['uploader_url'] = urljoin(
3203 'https://www.youtube.com/',
3204 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3205 return {k: v for k, v in uploader.items() if v is not None}
3206
3207 def _extract_from_tabs(self, item_id, webpage, data, tabs):
3208 playlist_id = title = description = channel_url = channel_name = channel_id = None
3209 thumbnails_list = tags = []
3210
3211 selected_tab = self._extract_selected_tab(tabs)
3212 renderer = try_get(
3213 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3214 if renderer:
3215 channel_name = renderer.get('title')
3216 channel_url = renderer.get('channelUrl')
3217 channel_id = renderer.get('externalId')
3218 else:
3219 renderer = try_get(
3220 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3221
3222 if renderer:
3223 title = renderer.get('title')
3224 description = renderer.get('description', '')
3225 playlist_id = channel_id
3226 tags = renderer.get('keywords', '').split()
3227 thumbnails_list = (
3228 try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
3229 or try_get(
3230 data,
3231 lambda x: x['sidebar']['playlistSidebarRenderer']['items'][0]['playlistSidebarPrimaryInfoRenderer']['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
3232 list)
3233 or [])
3234
3235 thumbnails = []
3236 for t in thumbnails_list:
3237 if not isinstance(t, dict):
3238 continue
3239 thumbnail_url = url_or_none(t.get('url'))
3240 if not thumbnail_url:
3241 continue
3242 thumbnails.append({
3243 'url': thumbnail_url,
3244 'width': int_or_none(t.get('width')),
3245 'height': int_or_none(t.get('height')),
3246 })
3247 if playlist_id is None:
3248 playlist_id = item_id
3249 if title is None:
3250 title = (
3251 try_get(data, lambda x: x['header']['hashtagHeaderRenderer']['hashtag']['simpleText'])
3252 or playlist_id)
3253 title += format_field(selected_tab, 'title', ' - %s')
3254
3255 metadata = {
3256 'playlist_id': playlist_id,
3257 'playlist_title': title,
3258 'playlist_description': description,
3259 'uploader': channel_name,
3260 'uploader_id': channel_id,
3261 'uploader_url': channel_url,
3262 'thumbnails': thumbnails,
3263 'tags': tags,
3264 }
3265 if not channel_id:
3266 metadata.update(self._extract_uploader(data))
3267 metadata.update({
3268 'channel': metadata['uploader'],
3269 'channel_id': metadata['uploader_id'],
3270 'channel_url': metadata['uploader_url']})
3271 return self.playlist_result(
3272 self._entries(
3273 selected_tab, playlist_id,
3274 self._extract_identity_token(webpage, item_id),
3275 self._extract_account_syncid(data),
3276 self._extract_ytcfg(item_id, webpage)),
3277 **metadata)
3278
3279 def _extract_mix_playlist(self, playlist, playlist_id, data, webpage):
3280 first_id = last_id = None
3281 ytcfg = self._extract_ytcfg(playlist_id, webpage)
3282 headers = self._generate_api_headers(
3283 ytcfg, account_syncid=self._extract_account_syncid(data),
3284 identity_token=self._extract_identity_token(webpage, item_id=playlist_id),
3285 visitor_data=try_get(self._extract_context(ytcfg), lambda x: x['client']['visitorData'], compat_str))
3286 for page_num in itertools.count(1):
3287 videos = list(self._playlist_entries(playlist))
3288 if not videos:
3289 return
3290 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
3291 if start >= len(videos):
3292 return
3293 for video in videos[start:]:
3294 if video['id'] == first_id:
3295 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
3296 return
3297 yield video
3298 first_id = first_id or videos[0]['id']
3299 last_id = videos[-1]['id']
3300 watch_endpoint = try_get(
3301 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
3302 query = {
3303 'playlistId': playlist_id,
3304 'videoId': watch_endpoint.get('videoId') or last_id,
3305 'index': watch_endpoint.get('index') or len(videos),
3306 'params': watch_endpoint.get('params') or 'OAE%3D'
3307 }
3308 response = self._extract_response(
3309 item_id='%s page %d' % (playlist_id, page_num),
3310 query=query,
3311 ep='next',
3312 headers=headers,
3313 check_get_keys='contents'
3314 )
3315 playlist = try_get(
3316 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3317
3318 def _extract_from_playlist(self, item_id, url, data, playlist, webpage):
3319 title = playlist.get('title') or try_get(
3320 data, lambda x: x['titleText']['simpleText'], compat_str)
3321 playlist_id = playlist.get('playlistId') or item_id
3322
3323 # Delegating everything except mix playlists to regular tab-based playlist URL
3324 playlist_url = urljoin(url, try_get(
3325 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3326 compat_str))
3327 if playlist_url and playlist_url != url:
3328 return self.url_result(
3329 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3330 video_title=title)
3331
3332 return self.playlist_result(
3333 self._extract_mix_playlist(playlist, playlist_id, data, webpage),
3334 playlist_id=playlist_id, playlist_title=title)
3335
3336 def _extract_alerts(self, data, expected=False):
3337
3338 def _real_extract_alerts():
3339 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
3340 if not isinstance(alert_dict, dict):
3341 continue
3342 for alert in alert_dict.values():
3343 alert_type = alert.get('type')
3344 if not alert_type:
3345 continue
3346 message = try_get(alert, lambda x: x['text']['simpleText'], compat_str) or ''
3347 if message:
3348 yield alert_type, message
3349 for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
3350 message += try_get(run, lambda x: x['text'], compat_str)
3351 if message:
3352 yield alert_type, message
3353
3354 errors = []
3355 warnings = []
3356 for alert_type, alert_message in _real_extract_alerts():
3357 if alert_type.lower() == 'error':
3358 errors.append([alert_type, alert_message])
3359 else:
3360 warnings.append([alert_type, alert_message])
3361
3362 for alert_type, alert_message in (warnings + errors[:-1]):
3363 self.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
3364 if errors:
3365 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
3366
3367 def _reload_with_unavailable_videos(self, item_id, data, webpage):
3368 """
3369 Get playlist with unavailable videos if the 'show unavailable videos' button exists.
3370 """
3371 sidebar_renderer = try_get(
3372 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3373 if not sidebar_renderer:
3374 return
3375 browse_id = params = None
3376 for item in sidebar_renderer:
3377 if not isinstance(item, dict):
3378 continue
3379 renderer = item.get('playlistSidebarPrimaryInfoRenderer')
3380 menu_renderer = try_get(
3381 renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
3382 for menu_item in menu_renderer:
3383 if not isinstance(menu_item, dict):
3384 continue
3385 nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
3386 text = try_get(
3387 nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
3388 if not text or text.lower() != 'show unavailable videos':
3389 continue
3390 browse_endpoint = try_get(
3391 nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {}
3392 browse_id = browse_endpoint.get('browseId')
3393 params = browse_endpoint.get('params')
3394 break
3395
3396 ytcfg = self._extract_ytcfg(item_id, webpage)
3397 headers = self._generate_api_headers(
3398 ytcfg, account_syncid=self._extract_account_syncid(ytcfg),
3399 identity_token=self._extract_identity_token(webpage, item_id=item_id),
3400 visitor_data=try_get(
3401 self._extract_context(ytcfg), lambda x: x['client']['visitorData'], compat_str))
3402 query = {
3403 'params': params or 'wgYCCAA=',
3404 'browseId': browse_id or 'VL%s' % item_id
3405 }
3406 return self._extract_response(
3407 item_id=item_id, headers=headers, query=query,
3408 check_get_keys='contents', fatal=False,
3409 note='Downloading API JSON with unavailable videos')
3410
3411 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
3412 ytcfg=None, check_get_keys=None, ep='browse', fatal=True):
3413 response = None
3414 last_error = None
3415 count = -1
3416 retries = self._downloader.params.get('extractor_retries', 3)
3417 if check_get_keys is None:
3418 check_get_keys = []
3419 while count < retries:
3420 count += 1
3421 if last_error:
3422 self.report_warning('%s. Retrying ...' % last_error)
3423 try:
3424 response = self._call_api(
3425 ep=ep, fatal=True, headers=headers,
3426 video_id=item_id, query=query,
3427 context=self._extract_context(ytcfg),
3428 api_key=self._extract_api_key(ytcfg),
3429 note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
3430 except ExtractorError as e:
3431 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404):
3432 # Downloading page may result in intermittent 5xx HTTP error
3433 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
3434 last_error = 'HTTP Error %s' % e.cause.code
3435 if count < retries:
3436 continue
3437 if fatal:
3438 raise
3439 else:
3440 self.report_warning(error_to_compat_str(e))
3441 return
3442
3443 else:
3444 # Youtube may send alerts if there was an issue with the continuation page
3445 self._extract_alerts(response, expected=False)
3446 if not check_get_keys or dict_get(response, check_get_keys):
3447 break
3448 # Youtube sometimes sends incomplete data
3449 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
3450 last_error = 'Incomplete data received'
3451 if count >= retries:
3452 if fatal:
3453 raise ExtractorError(last_error)
3454 else:
3455 self.report_warning(last_error)
3456 return
3457 return response
3458
3459 def _extract_webpage(self, url, item_id):
3460 retries = self._downloader.params.get('extractor_retries', 3)
3461 count = -1
3462 last_error = 'Incomplete yt initial data recieved'
3463 while count < retries:
3464 count += 1
3465 # Sometimes youtube returns a webpage with incomplete ytInitialData
3466 # See: https://github.com/yt-dlp/yt-dlp/issues/116
3467 if count:
3468 self.report_warning('%s. Retrying ...' % last_error)
3469 webpage = self._download_webpage(
3470 url, item_id,
3471 'Downloading webpage%s' % (' (retry #%d)' % count if count else ''))
3472 data = self._extract_yt_initial_data(item_id, webpage)
3473 self._extract_alerts(data, expected=True)
3474 if data.get('contents') or data.get('currentVideoEndpoint'):
3475 break
3476 if count >= retries:
3477 raise ExtractorError(last_error)
3478 return webpage, data
3479
3480 def _real_extract(self, url):
3481 item_id = self._match_id(url)
3482 url = compat_urlparse.urlunparse(
3483 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
3484
3485 # This is not matched in a channel page with a tab selected
3486 mobj = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
3487 mobj = mobj.groupdict() if mobj else {}
3488 if mobj and not mobj.get('not_channel'):
3489 self.report_warning(
3490 'A channel/user page was given. All the channel\'s videos will be downloaded. '
3491 'To download only the videos in the home page, add a "/featured" to the URL')
3492 url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '')
3493
3494 # Handle both video/playlist URLs
3495 qs = parse_qs(url)
3496 video_id = qs.get('v', [None])[0]
3497 playlist_id = qs.get('list', [None])[0]
3498
3499 if not video_id and (mobj.get('not_channel') or '').startswith('watch'):
3500 if not playlist_id:
3501 # If there is neither video or playlist ids,
3502 # youtube redirects to home page, which is undesirable
3503 raise ExtractorError('Unable to recognize tab page')
3504 self.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
3505 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
3506
3507 if video_id and playlist_id:
3508 if self._downloader.params.get('noplaylist'):
3509 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3510 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3511 self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
3512
3513 webpage, data = self._extract_webpage(url, item_id)
3514
3515 # YouTube sometimes provides a button to reload playlist with unavailable videos.
3516 data = self._reload_with_unavailable_videos(item_id, data, webpage) or data
3517
3518 tabs = try_get(
3519 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
3520 if tabs:
3521 return self._extract_from_tabs(item_id, webpage, data, tabs)
3522
3523 playlist = try_get(
3524 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3525 if playlist:
3526 return self._extract_from_playlist(item_id, url, data, playlist, webpage)
3527
3528 video_id = try_get(
3529 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
3530 compat_str) or video_id
3531 if video_id:
3532 self.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
3533 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3534
3535 raise ExtractorError('Unable to recognize tab page')
3536
3537
3538 class YoutubePlaylistIE(InfoExtractor):
3539 IE_DESC = 'YouTube.com playlists'
3540 _VALID_URL = r'''(?x)(?:
3541 (?:https?://)?
3542 (?:\w+\.)?
3543 (?:
3544 (?:
3545 youtube(?:kids)?\.com|
3546 invidio\.us
3547 )
3548 /.*?\?.*?\blist=
3549 )?
3550 (?P<id>%(playlist_id)s)
3551 )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3552 IE_NAME = 'youtube:playlist'
3553 _TESTS = [{
3554 'note': 'issue #673',
3555 'url': 'PLBB231211A4F62143',
3556 'info_dict': {
3557 'title': '[OLD]Team Fortress 2 (Class-based LP)',
3558 'id': 'PLBB231211A4F62143',
3559 'uploader': 'Wickydoo',
3560 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
3561 },
3562 'playlist_mincount': 29,
3563 }, {
3564 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3565 'info_dict': {
3566 'title': 'YDL_safe_search',
3567 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3568 },
3569 'playlist_count': 2,
3570 'skip': 'This playlist is private',
3571 }, {
3572 'note': 'embedded',
3573 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3574 'playlist_count': 4,
3575 'info_dict': {
3576 'title': 'JODA15',
3577 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3578 'uploader': 'milan',
3579 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
3580 }
3581 }, {
3582 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3583 'playlist_mincount': 982,
3584 'info_dict': {
3585 'title': '2018 Chinese New Singles (11/6 updated)',
3586 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3587 'uploader': 'LBK',
3588 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
3589 }
3590 }, {
3591 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
3592 'only_matching': True,
3593 }, {
3594 # music album playlist
3595 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
3596 'only_matching': True,
3597 }]
3598
3599 @classmethod
3600 def suitable(cls, url):
3601 if YoutubeTabIE.suitable(url):
3602 return False
3603 # Hack for lazy extractors until more generic solution is implemented
3604 # (see #28780)
3605 from .youtube import parse_qs
3606 qs = parse_qs(url)
3607 if qs.get('v', [None])[0]:
3608 return False
3609 return super(YoutubePlaylistIE, cls).suitable(url)
3610
3611 def _real_extract(self, url):
3612 playlist_id = self._match_id(url)
3613 qs = parse_qs(url)
3614 if not qs:
3615 qs = {'list': playlist_id}
3616 return self.url_result(
3617 update_url_query('https://www.youtube.com/playlist', qs),
3618 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3619
3620
3621 class YoutubeYtBeIE(InfoExtractor):
3622 IE_DESC = 'youtu.be'
3623 _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}
3624 _TESTS = [{
3625 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
3626 'info_dict': {
3627 'id': 'yeWKywCrFtk',
3628 'ext': 'mp4',
3629 'title': 'Small Scale Baler and Braiding Rugs',
3630 'uploader': 'Backus-Page House Museum',
3631 'uploader_id': 'backuspagemuseum',
3632 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
3633 'upload_date': '20161008',
3634 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
3635 'categories': ['Nonprofits & Activism'],
3636 'tags': list,
3637 'like_count': int,
3638 'dislike_count': int,
3639 },
3640 'params': {
3641 'noplaylist': True,
3642 'skip_download': True,
3643 },
3644 }, {
3645 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
3646 'only_matching': True,
3647 }]
3648
3649 def _real_extract(self, url):
3650 mobj = re.match(self._VALID_URL, url)
3651 video_id = mobj.group('id')
3652 playlist_id = mobj.group('playlist_id')
3653 return self.url_result(
3654 update_url_query('https://www.youtube.com/watch', {
3655 'v': video_id,
3656 'list': playlist_id,
3657 'feature': 'youtu.be',
3658 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3659
3660
3661 class YoutubeYtUserIE(InfoExtractor):
3662 IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword'
3663 _VALID_URL = r'ytuser:(?P<id>.+)'
3664 _TESTS = [{
3665 'url': 'ytuser:phihag',
3666 'only_matching': True,
3667 }]
3668
3669 def _real_extract(self, url):
3670 user_id = self._match_id(url)
3671 return self.url_result(
3672 'https://www.youtube.com/user/%s' % user_id,
3673 ie=YoutubeTabIE.ie_key(), video_id=user_id)
3674
3675
3676 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3677 IE_NAME = 'youtube:favorites'
3678 IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)'
3679 _VALID_URL = r':ytfav(?:ou?rite)?s?'
3680 _LOGIN_REQUIRED = True
3681 _TESTS = [{
3682 'url': ':ytfav',
3683 'only_matching': True,
3684 }, {
3685 'url': ':ytfavorites',
3686 'only_matching': True,
3687 }]
3688
3689 def _real_extract(self, url):
3690 return self.url_result(
3691 'https://www.youtube.com/playlist?list=LL',
3692 ie=YoutubeTabIE.ie_key())
3693
3694
3695 class YoutubeSearchIE(SearchInfoExtractor, YoutubeTabIE):
3696 IE_DESC = 'YouTube.com searches, "ytsearch" keyword'
3697 # there doesn't appear to be a real limit, for example if you search for
3698 # 'python' you get more than 8.000.000 results
3699 _MAX_RESULTS = float('inf')
3700 IE_NAME = 'youtube:search'
3701 _SEARCH_KEY = 'ytsearch'
3702 _SEARCH_PARAMS = None
3703 _TESTS = []
3704
3705 def _entries(self, query, n):
3706 data = {'query': query}
3707 if self._SEARCH_PARAMS:
3708 data['params'] = self._SEARCH_PARAMS
3709 total = 0
3710 for page_num in itertools.count(1):
3711 search = self._extract_response(
3712 item_id='query "%s" page %s' % (query, page_num), ep='search', query=data,
3713 check_get_keys=('contents', 'onResponseReceivedCommands')
3714 )
3715 if not search:
3716 break
3717 slr_contents = try_get(
3718 search,
3719 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
3720 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
3721 list)
3722 if not slr_contents:
3723 break
3724
3725 # Youtube sometimes adds promoted content to searches,
3726 # changing the index location of videos and token.
3727 # So we search through all entries till we find them.
3728 continuation_token = None
3729 for slr_content in slr_contents:
3730 if continuation_token is None:
3731 continuation_token = try_get(
3732 slr_content,
3733 lambda x: x['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'],
3734 compat_str)
3735
3736 isr_contents = try_get(
3737 slr_content,
3738 lambda x: x['itemSectionRenderer']['contents'],
3739 list)
3740 if not isr_contents:
3741 continue
3742 for content in isr_contents:
3743 if not isinstance(content, dict):
3744 continue
3745 video = content.get('videoRenderer')
3746 if not isinstance(video, dict):
3747 continue
3748 video_id = video.get('videoId')
3749 if not video_id:
3750 continue
3751
3752 yield self._extract_video(video)
3753 total += 1
3754 if total == n:
3755 return
3756
3757 if not continuation_token:
3758 break
3759 data['continuation'] = continuation_token
3760
3761 def _get_n_results(self, query, n):
3762 """Get a specified number of results for a query"""
3763 return self.playlist_result(self._entries(query, n), query)
3764
3765
3766 class YoutubeSearchDateIE(YoutubeSearchIE):
3767 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3768 _SEARCH_KEY = 'ytsearchdate'
3769 IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword'
3770 _SEARCH_PARAMS = 'CAI%3D'
3771
3772
3773 class YoutubeSearchURLIE(YoutubeSearchIE):
3774 IE_DESC = 'YouTube.com search URLs'
3775 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
3776 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
3777 # _MAX_RESULTS = 100
3778 _TESTS = [{
3779 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3780 'playlist_mincount': 5,
3781 'info_dict': {
3782 'title': 'youtube-dl test video',
3783 }
3784 }, {
3785 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3786 'only_matching': True,
3787 }]
3788
3789 @classmethod
3790 def _make_valid_url(cls):
3791 return cls._VALID_URL
3792
3793 def _real_extract(self, url):
3794 qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
3795 query = (qs.get('search_query') or qs.get('q'))[0]
3796 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
3797 return self._get_n_results(query, self._MAX_RESULTS)
3798
3799
3800 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
3801 """
3802 Base class for feed extractors
3803 Subclasses must define the _FEED_NAME property.
3804 """
3805 _LOGIN_REQUIRED = True
3806 _TESTS = []
3807
3808 @property
3809 def IE_NAME(self):
3810 return 'youtube:%s' % self._FEED_NAME
3811
3812 def _real_initialize(self):
3813 self._login()
3814
3815 def _real_extract(self, url):
3816 return self.url_result(
3817 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3818 ie=YoutubeTabIE.ie_key())
3819
3820
3821 class YoutubeWatchLaterIE(InfoExtractor):
3822 IE_NAME = 'youtube:watchlater'
3823 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3824 _VALID_URL = r':ytwatchlater'
3825 _TESTS = [{
3826 'url': ':ytwatchlater',
3827 'only_matching': True,
3828 }]
3829
3830 def _real_extract(self, url):
3831 return self.url_result(
3832 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
3833
3834
3835 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3836 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3837 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
3838 _FEED_NAME = 'recommended'
3839 _TESTS = [{
3840 'url': ':ytrec',
3841 'only_matching': True,
3842 }, {
3843 'url': ':ytrecommended',
3844 'only_matching': True,
3845 }, {
3846 'url': 'https://youtube.com',
3847 'only_matching': True,
3848 }]
3849
3850
3851 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3852 IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)'
3853 _VALID_URL = r':ytsub(?:scription)?s?'
3854 _FEED_NAME = 'subscriptions'
3855 _TESTS = [{
3856 'url': ':ytsubs',
3857 'only_matching': True,
3858 }, {
3859 'url': ':ytsubscriptions',
3860 'only_matching': True,
3861 }]
3862
3863
3864 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3865 IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)'
3866 _VALID_URL = r':ythis(?:tory)?'
3867 _FEED_NAME = 'history'
3868 _TESTS = [{
3869 'url': ':ythistory',
3870 'only_matching': True,
3871 }]
3872
3873
3874 class YoutubeTruncatedURLIE(InfoExtractor):
3875 IE_NAME = 'youtube:truncated_url'
3876 IE_DESC = False # Do not list
3877 _VALID_URL = r'''(?x)
3878 (?:https?://)?
3879 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3880 (?:watch\?(?:
3881 feature=[a-z_]+|
3882 annotation_id=annotation_[^&]+|
3883 x-yt-cl=[0-9]+|
3884 hl=[^&]*|
3885 t=[0-9]+
3886 )?
3887 |
3888 attribution_link\?a=[^&]+
3889 )
3890 $
3891 '''
3892
3893 _TESTS = [{
3894 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3895 'only_matching': True,
3896 }, {
3897 'url': 'https://www.youtube.com/watch?',
3898 'only_matching': True,
3899 }, {
3900 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3901 'only_matching': True,
3902 }, {
3903 'url': 'https://www.youtube.com/watch?feature=foo',
3904 'only_matching': True,
3905 }, {
3906 'url': 'https://www.youtube.com/watch?hl=en-GB',
3907 'only_matching': True,
3908 }, {
3909 'url': 'https://www.youtube.com/watch?t=2372',
3910 'only_matching': True,
3911 }]
3912
3913 def _real_extract(self, url):
3914 raise ExtractorError(
3915 'Did you forget to quote the URL? Remember that & is a meta '
3916 'character in most shells, so you want to put the URL in quotes, '
3917 'like youtube-dl '
3918 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3919 ' or simply youtube-dl BaW_jenozKc .',
3920 expected=True)
3921
3922
3923 class YoutubeTruncatedIDIE(InfoExtractor):
3924 IE_NAME = 'youtube:truncated_id'
3925 IE_DESC = False # Do not list
3926 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3927
3928 _TESTS = [{
3929 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3930 'only_matching': True,
3931 }]
3932
3933 def _real_extract(self, url):
3934 video_id = self._match_id(url)
3935 raise ExtractorError(
3936 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3937 expected=True)
fork of https://github.com/yt-dlp/yt-dlp/