]> 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
Update to ytdl-commit-4fb25ff
[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 ExtractorError,
33 format_field,
34 float_or_none,
35 int_or_none,
36 mimetype2ext,
37 parse_codecs,
38 parse_duration,
39 qualities,
40 remove_start,
41 smuggle_url,
42 str_or_none,
43 str_to_int,
44 try_get,
45 unescapeHTML,
46 unified_strdate,
47 unsmuggle_url,
48 update_url_query,
49 url_or_none,
50 urlencode_postdata,
51 urljoin
52 )
53
54
55 class YoutubeBaseInfoExtractor(InfoExtractor):
56 """Provide base functions for Youtube extractors"""
57 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
58 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
59
60 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
61 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
62 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
63
64 _RESERVED_NAMES = (
65 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|'
66 r'movies|results|shared|hashtag|trending|feed|feeds|'
67 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
68
69 _NETRC_MACHINE = 'youtube'
70 # If True it will raise an error if no login info is provided
71 _LOGIN_REQUIRED = False
72
73 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
74
75 def _ids_to_results(self, ids):
76 return [
77 self.url_result(vid_id, 'Youtube', video_id=vid_id)
78 for vid_id in ids]
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._downloader.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 = '2.20210301.08.00'
288 _DEFAULT_API_DATA = {
289 'context': {
290 'client': {
291 'clientName': 'WEB',
292 'clientVersion': _YT_WEB_CLIENT_VERSION,
293 }
294 },
295 }
296
297 _DEFAULT_BASIC_API_HEADERS = {
298 'X-YouTube-Client-Name': '1',
299 'X-YouTube-Client-Version': _YT_WEB_CLIENT_VERSION
300 }
301
302 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
303 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
304 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
305
306 def _generate_sapisidhash_header(self):
307 sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
308 if sapisid_cookie is None:
309 return
310 time_now = round(time.time())
311 sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
312 return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
313
314 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
315 note='Downloading API JSON', errnote='Unable to download API page'):
316 data = self._DEFAULT_API_DATA.copy()
317 data.update(query)
318 headers = headers or {}
319 headers.update({'content-type': 'application/json'})
320 auth = self._generate_sapisidhash_header()
321 if auth is not None:
322 headers.update({'Authorization': auth, 'X-Origin': 'https://www.youtube.com'})
323 return self._download_json(
324 'https://www.youtube.com/youtubei/v1/%s' % ep,
325 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
326 data=json.dumps(data).encode('utf8'), headers=headers,
327 query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'})
328
329 def _extract_yt_initial_data(self, video_id, webpage):
330 return self._parse_json(
331 self._search_regex(
332 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
333 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
334 video_id)
335
336 def _extract_identity_token(self, webpage, item_id):
337 ytcfg = self._extract_ytcfg(item_id, webpage)
338 if ytcfg:
339 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
340 if token:
341 return token
342 return self._search_regex(
343 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
344 'identity token', default=None)
345
346 @staticmethod
347 def _extract_account_syncid(data):
348 """Extract syncId required to download private playlists of secondary channels"""
349 sync_ids = (
350 try_get(data, lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'], compat_str)
351 or '').split("||")
352 if len(sync_ids) >= 2 and sync_ids[1]:
353 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
354 # and just "user_syncid||" for primary channel. We only want the channel_syncid
355 return sync_ids[0]
356
357 def _extract_ytcfg(self, video_id, webpage):
358 return self._parse_json(
359 self._search_regex(
360 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
361 default='{}'), video_id, fatal=False)
362
363 def _extract_video(self, renderer):
364 video_id = renderer.get('videoId')
365 title = try_get(
366 renderer,
367 (lambda x: x['title']['runs'][0]['text'],
368 lambda x: x['title']['simpleText']), compat_str)
369 description = try_get(
370 renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'],
371 compat_str)
372 duration = parse_duration(try_get(
373 renderer, lambda x: x['lengthText']['simpleText'], compat_str))
374 view_count_text = try_get(
375 renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
376 view_count = str_to_int(self._search_regex(
377 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
378 'view count', default=None))
379 uploader = try_get(
380 renderer,
381 (lambda x: x['ownerText']['runs'][0]['text'],
382 lambda x: x['shortBylineText']['runs'][0]['text']), compat_str)
383 return {
384 '_type': 'url',
385 'ie_key': YoutubeIE.ie_key(),
386 'id': video_id,
387 'url': video_id,
388 'title': title,
389 'description': description,
390 'duration': duration,
391 'view_count': view_count,
392 'uploader': uploader,
393 }
394
395
396 class YoutubeIE(YoutubeBaseInfoExtractor):
397 IE_DESC = 'YouTube.com'
398 _INVIDIOUS_SITES = (
399 # invidious-redirect websites
400 r'(?:www\.)?redirect\.invidious\.io',
401 r'(?:(?:www|dev)\.)?invidio\.us',
402 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
403 r'(?:www\.)?invidious\.pussthecat\.org',
404 r'(?:www\.)?invidious\.048596\.xyz',
405 r'(?:www\.)?invidious\.zee\.li',
406 r'(?:www\.)?vid\.puffyan\.us',
407 r'(?:(?:www|au)\.)?ytprivate\.com',
408 r'(?:www\.)?invidious\.namazso\.eu',
409 r'(?:www\.)?invidious\.ethibox\.fr',
410 r'(?:www\.)?inv\.skyn3t\.in',
411 r'(?:www\.)?invidious\.himiko\.cloud',
412 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
413 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
414 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
415 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
416 # youtube-dl invidious instances list
417 r'(?:(?:www|no)\.)?invidiou\.sh',
418 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
419 r'(?:www\.)?invidious\.kabi\.tk',
420 r'(?:www\.)?invidious\.13ad\.de',
421 r'(?:www\.)?invidious\.mastodon\.host',
422 r'(?:www\.)?invidious\.zapashcanon\.fr',
423 r'(?:www\.)?invidious\.kavin\.rocks',
424 r'(?:www\.)?invidious\.tube',
425 r'(?:www\.)?invidiou\.site',
426 r'(?:www\.)?invidious\.site',
427 r'(?:www\.)?invidious\.xyz',
428 r'(?:www\.)?invidious\.nixnet\.xyz',
429 r'(?:www\.)?invidious\.drycat\.fr',
430 r'(?:www\.)?tube\.poal\.co',
431 r'(?:www\.)?tube\.connect\.cafe',
432 r'(?:www\.)?vid\.wxzm\.sx',
433 r'(?:www\.)?vid\.mint\.lgbt',
434 r'(?:www\.)?yewtu\.be',
435 r'(?:www\.)?yt\.elukerio\.org',
436 r'(?:www\.)?yt\.lelux\.fi',
437 r'(?:www\.)?invidious\.ggc-project\.de',
438 r'(?:www\.)?yt\.maisputain\.ovh',
439 r'(?:www\.)?invidious\.toot\.koeln',
440 r'(?:www\.)?invidious\.fdn\.fr',
441 r'(?:www\.)?watch\.nettohikari\.com',
442 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
443 r'(?:www\.)?qklhadlycap4cnod\.onion',
444 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
445 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
446 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
447 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
448 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
449 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
450 )
451 _VALID_URL = r"""(?x)^
452 (
453 (?:https?://|//) # http(s):// or protocol-independent URL
454 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
455 (?:www\.)?deturl\.com/www\.youtube\.com|
456 (?:www\.)?pwnyoutube\.com|
457 (?:www\.)?hooktube\.com|
458 (?:www\.)?yourepeat\.com|
459 tube\.majestyc\.net|
460 %(invidious)s|
461 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
462 (?:.*?\#/)? # handle anchor (#/) redirect urls
463 (?: # the various things that can precede the ID:
464 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
465 |(?: # or the v= param in all its forms
466 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
467 (?:\?|\#!?) # the params delimiter ? or # or #!
468 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
469 v=
470 )
471 ))
472 |(?:
473 youtu\.be| # just youtu.be/xxxx
474 vid\.plus| # or vid.plus/xxxx
475 zwearz\.com/watch| # or zwearz.com/watch/xxxx
476 %(invidious)s
477 )/
478 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
479 )
480 )? # all until now is optional -> you can pass the naked ID
481 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
482 (?!.*?\blist=
483 (?:
484 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
485 WL # WL are handled by the watch later IE
486 )
487 )
488 (?(1).+)? # if we found the ID, everything can follow
489 $""" % {
490 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
491 'invidious': '|'.join(_INVIDIOUS_SITES),
492 }
493 _PLAYER_INFO_RE = (
494 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
495 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
496 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
497 )
498 _formats = {
499 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
500 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
501 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
502 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
503 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
504 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
505 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
506 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
507 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
508 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
509 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
510 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
511 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
512 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
513 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
514 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
515 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
516 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
517
518
519 # 3D videos
520 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
521 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
522 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
523 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
524 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
525 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
526 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
527
528 # Apple HTTP Live Streaming
529 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
530 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
531 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
532 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
533 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
534 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
535 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
536 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
537
538 # DASH mp4 video
539 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
540 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
541 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
542 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
543 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
544 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
545 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
546 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
547 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
548 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
549 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
550 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
551
552 # Dash mp4 audio
553 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
554 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
555 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
556 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
557 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
558 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
559 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
560
561 # Dash webm
562 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
563 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
564 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
565 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
566 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
567 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
568 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
569 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
570 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
571 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
572 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
573 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
574 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
575 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
576 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
577 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
578 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
579 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
580 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
581 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
582 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
583 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
584
585 # Dash webm audio
586 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
587 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
588
589 # Dash webm audio with opus inside
590 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
591 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
592 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
593
594 # RTMP (unnamed)
595 '_rtmp': {'protocol': 'rtmp'},
596
597 # av01 video only formats sometimes served with "unknown" codecs
598 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
599 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
600 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
601 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
602 }
603 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
604
605 _GEO_BYPASS = False
606
607 IE_NAME = 'youtube'
608 _TESTS = [
609 {
610 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
611 'info_dict': {
612 'id': 'BaW_jenozKc',
613 'ext': 'mp4',
614 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
615 'uploader': 'Philipp Hagemeister',
616 'uploader_id': 'phihag',
617 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
618 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
619 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
620 'upload_date': '20121002',
621 '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 .',
622 'categories': ['Science & Technology'],
623 'tags': ['youtube-dl'],
624 'duration': 10,
625 'view_count': int,
626 'like_count': int,
627 'dislike_count': int,
628 'start_time': 1,
629 'end_time': 9,
630 }
631 },
632 {
633 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
634 'note': 'Embed-only video (#1746)',
635 'info_dict': {
636 'id': 'yZIXLfi8CZQ',
637 'ext': 'mp4',
638 'upload_date': '20120608',
639 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
640 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
641 'uploader': 'SET India',
642 'uploader_id': 'setindia',
643 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
644 'age_limit': 18,
645 },
646 'skip': 'Private video',
647 },
648 {
649 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
650 'note': 'Use the first video ID in the URL',
651 'info_dict': {
652 'id': 'BaW_jenozKc',
653 'ext': 'mp4',
654 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
655 'uploader': 'Philipp Hagemeister',
656 'uploader_id': 'phihag',
657 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
658 'upload_date': '20121002',
659 '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 .',
660 'categories': ['Science & Technology'],
661 'tags': ['youtube-dl'],
662 'duration': 10,
663 'view_count': int,
664 'like_count': int,
665 'dislike_count': int,
666 },
667 'params': {
668 'skip_download': True,
669 },
670 },
671 {
672 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
673 'note': '256k DASH audio (format 141) via DASH manifest',
674 'info_dict': {
675 'id': 'a9LDPn-MO4I',
676 'ext': 'm4a',
677 'upload_date': '20121002',
678 'uploader_id': '8KVIDEO',
679 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
680 'description': '',
681 'uploader': '8KVIDEO',
682 'title': 'UHDTV TEST 8K VIDEO.mp4'
683 },
684 'params': {
685 'youtube_include_dash_manifest': True,
686 'format': '141',
687 },
688 'skip': 'format 141 not served anymore',
689 },
690 # DASH manifest with encrypted signature
691 {
692 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
693 'info_dict': {
694 'id': 'IB3lcPjvWLA',
695 'ext': 'm4a',
696 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
697 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
698 'duration': 244,
699 'uploader': 'AfrojackVEVO',
700 'uploader_id': 'AfrojackVEVO',
701 'upload_date': '20131011',
702 'abr': 129.495,
703 },
704 'params': {
705 'youtube_include_dash_manifest': True,
706 'format': '141/bestaudio[ext=m4a]',
707 },
708 },
709 # Controversy video
710 {
711 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
712 'info_dict': {
713 'id': 'T4XJQO3qol8',
714 'ext': 'mp4',
715 'duration': 219,
716 'upload_date': '20100909',
717 'uploader': 'Amazing Atheist',
718 'uploader_id': 'TheAmazingAtheist',
719 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
720 'title': 'Burning Everyone\'s Koran',
721 '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',
722 }
723 },
724 # Normal age-gate video (embed allowed)
725 {
726 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
727 'info_dict': {
728 'id': 'HtVdAasjOgU',
729 'ext': 'mp4',
730 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
731 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
732 'duration': 142,
733 'uploader': 'The Witcher',
734 'uploader_id': 'WitcherGame',
735 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
736 'upload_date': '20140605',
737 'age_limit': 18,
738 },
739 },
740 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
741 # YouTube Red ad is not captured for creator
742 {
743 'url': '__2ABJjxzNo',
744 'info_dict': {
745 'id': '__2ABJjxzNo',
746 'ext': 'mp4',
747 'duration': 266,
748 'upload_date': '20100430',
749 'uploader_id': 'deadmau5',
750 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
751 'creator': 'deadmau5',
752 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
753 'uploader': 'deadmau5',
754 'title': 'Deadmau5 - Some Chords (HD)',
755 'alt_title': 'Some Chords',
756 },
757 'expected_warnings': [
758 'DASH manifest missing',
759 ]
760 },
761 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
762 {
763 'url': 'lqQg6PlCWgI',
764 'info_dict': {
765 'id': 'lqQg6PlCWgI',
766 'ext': 'mp4',
767 'duration': 6085,
768 'upload_date': '20150827',
769 'uploader_id': 'olympic',
770 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
771 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
772 'uploader': 'Olympic',
773 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
774 },
775 'params': {
776 'skip_download': 'requires avconv',
777 }
778 },
779 # Non-square pixels
780 {
781 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
782 'info_dict': {
783 'id': '_b-2C3KPAM0',
784 'ext': 'mp4',
785 'stretched_ratio': 16 / 9.,
786 'duration': 85,
787 'upload_date': '20110310',
788 'uploader_id': 'AllenMeow',
789 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
790 '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',
791 'uploader': '孫ᄋᄅ',
792 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
793 },
794 },
795 # url_encoded_fmt_stream_map is empty string
796 {
797 'url': 'qEJwOuvDf7I',
798 'info_dict': {
799 'id': 'qEJwOuvDf7I',
800 'ext': 'webm',
801 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
802 'description': '',
803 'upload_date': '20150404',
804 'uploader_id': 'spbelect',
805 'uploader': 'Наблюдатели Петербурга',
806 },
807 'params': {
808 'skip_download': 'requires avconv',
809 },
810 'skip': 'This live event has ended.',
811 },
812 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
813 {
814 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
815 'info_dict': {
816 'id': 'FIl7x6_3R5Y',
817 'ext': 'webm',
818 'title': 'md5:7b81415841e02ecd4313668cde88737a',
819 'description': 'md5:116377fd2963b81ec4ce64b542173306',
820 'duration': 220,
821 'upload_date': '20150625',
822 'uploader_id': 'dorappi2000',
823 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
824 'uploader': 'dorappi2000',
825 'formats': 'mincount:31',
826 },
827 'skip': 'not actual anymore',
828 },
829 # DASH manifest with segment_list
830 {
831 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
832 'md5': '8ce563a1d667b599d21064e982ab9e31',
833 'info_dict': {
834 'id': 'CsmdDsKjzN8',
835 'ext': 'mp4',
836 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
837 'uploader': 'Airtek',
838 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
839 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
840 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
841 },
842 'params': {
843 'youtube_include_dash_manifest': True,
844 'format': '135', # bestvideo
845 },
846 'skip': 'This live event has ended.',
847 },
848 {
849 # Multifeed videos (multiple cameras), URL is for Main Camera
850 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
851 'info_dict': {
852 'id': 'jvGDaLqkpTg',
853 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
854 'description': 'md5:e03b909557865076822aa169218d6a5d',
855 },
856 'playlist': [{
857 'info_dict': {
858 'id': 'jvGDaLqkpTg',
859 'ext': 'mp4',
860 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
861 'description': 'md5:e03b909557865076822aa169218d6a5d',
862 'duration': 10643,
863 'upload_date': '20161111',
864 'uploader': 'Team PGP',
865 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
866 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
867 },
868 }, {
869 'info_dict': {
870 'id': '3AKt1R1aDnw',
871 'ext': 'mp4',
872 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
873 'description': 'md5:e03b909557865076822aa169218d6a5d',
874 'duration': 10991,
875 'upload_date': '20161111',
876 'uploader': 'Team PGP',
877 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
878 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
879 },
880 }, {
881 'info_dict': {
882 'id': 'RtAMM00gpVc',
883 'ext': 'mp4',
884 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
885 'description': 'md5:e03b909557865076822aa169218d6a5d',
886 'duration': 10995,
887 'upload_date': '20161111',
888 'uploader': 'Team PGP',
889 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
890 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
891 },
892 }, {
893 'info_dict': {
894 'id': '6N2fdlP3C5U',
895 'ext': 'mp4',
896 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
897 'description': 'md5:e03b909557865076822aa169218d6a5d',
898 'duration': 10990,
899 'upload_date': '20161111',
900 'uploader': 'Team PGP',
901 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
902 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
903 },
904 }],
905 'params': {
906 'skip_download': True,
907 },
908 },
909 {
910 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
911 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
912 'info_dict': {
913 'id': 'gVfLd0zydlo',
914 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
915 },
916 'playlist_count': 2,
917 'skip': 'Not multifeed anymore',
918 },
919 {
920 'url': 'https://vid.plus/FlRa-iH7PGw',
921 'only_matching': True,
922 },
923 {
924 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
925 'only_matching': True,
926 },
927 {
928 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
929 # Also tests cut-off URL expansion in video description (see
930 # https://github.com/ytdl-org/youtube-dl/issues/1892,
931 # https://github.com/ytdl-org/youtube-dl/issues/8164)
932 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
933 'info_dict': {
934 'id': 'lsguqyKfVQg',
935 'ext': 'mp4',
936 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
937 'alt_title': 'Dark Walk - Position Music',
938 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
939 'duration': 133,
940 'upload_date': '20151119',
941 'uploader_id': 'IronSoulElf',
942 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
943 'uploader': 'IronSoulElf',
944 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
945 'track': 'Dark Walk - Position Music',
946 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
947 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
948 },
949 'params': {
950 'skip_download': True,
951 },
952 },
953 {
954 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
955 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
956 'only_matching': True,
957 },
958 {
959 # Video with yt:stretch=17:0
960 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
961 'info_dict': {
962 'id': 'Q39EVAstoRM',
963 'ext': 'mp4',
964 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
965 'description': 'md5:ee18a25c350637c8faff806845bddee9',
966 'upload_date': '20151107',
967 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
968 'uploader': 'CH GAMER DROID',
969 },
970 'params': {
971 'skip_download': True,
972 },
973 'skip': 'This video does not exist.',
974 },
975 {
976 # Video licensed under Creative Commons
977 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
978 'info_dict': {
979 'id': 'M4gD1WSo5mA',
980 'ext': 'mp4',
981 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
982 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
983 'duration': 721,
984 'upload_date': '20150127',
985 'uploader_id': 'BerkmanCenter',
986 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
987 'uploader': 'The Berkman Klein Center for Internet & Society',
988 'license': 'Creative Commons Attribution license (reuse allowed)',
989 },
990 'params': {
991 'skip_download': True,
992 },
993 },
994 {
995 # Channel-like uploader_url
996 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
997 'info_dict': {
998 'id': 'eQcmzGIKrzg',
999 'ext': 'mp4',
1000 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1001 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1002 'duration': 4060,
1003 'upload_date': '20151119',
1004 'uploader': 'Bernie Sanders',
1005 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1006 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1007 'license': 'Creative Commons Attribution license (reuse allowed)',
1008 },
1009 'params': {
1010 'skip_download': True,
1011 },
1012 },
1013 {
1014 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1015 'only_matching': True,
1016 },
1017 {
1018 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1019 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1020 'only_matching': True,
1021 },
1022 {
1023 # Rental video preview
1024 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1025 'info_dict': {
1026 'id': 'uGpuVWrhIzE',
1027 'ext': 'mp4',
1028 'title': 'Piku - Trailer',
1029 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1030 'upload_date': '20150811',
1031 'uploader': 'FlixMatrix',
1032 'uploader_id': 'FlixMatrixKaravan',
1033 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1034 'license': 'Standard YouTube License',
1035 },
1036 'params': {
1037 'skip_download': True,
1038 },
1039 'skip': 'This video is not available.',
1040 },
1041 {
1042 # YouTube Red video with episode data
1043 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1044 'info_dict': {
1045 'id': 'iqKdEhx-dD4',
1046 'ext': 'mp4',
1047 'title': 'Isolation - Mind Field (Ep 1)',
1048 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1049 'duration': 2085,
1050 'upload_date': '20170118',
1051 'uploader': 'Vsauce',
1052 'uploader_id': 'Vsauce',
1053 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1054 'series': 'Mind Field',
1055 'season_number': 1,
1056 'episode_number': 1,
1057 },
1058 'params': {
1059 'skip_download': True,
1060 },
1061 'expected_warnings': [
1062 'Skipping DASH manifest',
1063 ],
1064 },
1065 {
1066 # The following content has been identified by the YouTube community
1067 # as inappropriate or offensive to some audiences.
1068 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1069 'info_dict': {
1070 'id': '6SJNVb0GnPI',
1071 'ext': 'mp4',
1072 'title': 'Race Differences in Intelligence',
1073 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1074 'duration': 965,
1075 'upload_date': '20140124',
1076 'uploader': 'New Century Foundation',
1077 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1078 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1079 },
1080 'params': {
1081 'skip_download': True,
1082 },
1083 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1084 },
1085 {
1086 # itag 212
1087 'url': '1t24XAntNCY',
1088 'only_matching': True,
1089 },
1090 {
1091 # geo restricted to JP
1092 'url': 'sJL6WA-aGkQ',
1093 'only_matching': True,
1094 },
1095 {
1096 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1097 'only_matching': True,
1098 },
1099 {
1100 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1101 'only_matching': True,
1102 },
1103 {
1104 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1105 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1106 'only_matching': True,
1107 },
1108 {
1109 # DRM protected
1110 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1111 'only_matching': True,
1112 },
1113 {
1114 # Video with unsupported adaptive stream type formats
1115 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1116 'info_dict': {
1117 'id': 'Z4Vy8R84T1U',
1118 'ext': 'mp4',
1119 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1120 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1121 'duration': 433,
1122 'upload_date': '20130923',
1123 'uploader': 'Amelia Putri Harwita',
1124 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1125 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1126 'formats': 'maxcount:10',
1127 },
1128 'params': {
1129 'skip_download': True,
1130 'youtube_include_dash_manifest': False,
1131 },
1132 'skip': 'not actual anymore',
1133 },
1134 {
1135 # Youtube Music Auto-generated description
1136 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1137 'info_dict': {
1138 'id': 'MgNrAu2pzNs',
1139 'ext': 'mp4',
1140 'title': 'Voyeur Girl',
1141 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1142 'upload_date': '20190312',
1143 'uploader': 'Stephen - Topic',
1144 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1145 'artist': 'Stephen',
1146 'track': 'Voyeur Girl',
1147 'album': 'it\'s too much love to know my dear',
1148 'release_date': '20190313',
1149 'release_year': 2019,
1150 },
1151 'params': {
1152 'skip_download': True,
1153 },
1154 },
1155 {
1156 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1157 'only_matching': True,
1158 },
1159 {
1160 # invalid -> valid video id redirection
1161 'url': 'DJztXj2GPfl',
1162 'info_dict': {
1163 'id': 'DJztXj2GPfk',
1164 'ext': 'mp4',
1165 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1166 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1167 'upload_date': '20090125',
1168 'uploader': 'Prochorowka',
1169 'uploader_id': 'Prochorowka',
1170 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1171 'artist': 'Panjabi MC',
1172 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1173 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1174 },
1175 'params': {
1176 'skip_download': True,
1177 },
1178 'skip': 'Video unavailable',
1179 },
1180 {
1181 # empty description results in an empty string
1182 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1183 'info_dict': {
1184 'id': 'x41yOUIvK2k',
1185 'ext': 'mp4',
1186 'title': 'IMG 3456',
1187 'description': '',
1188 'upload_date': '20170613',
1189 'uploader_id': 'ElevageOrVert',
1190 'uploader': 'ElevageOrVert',
1191 },
1192 'params': {
1193 'skip_download': True,
1194 },
1195 },
1196 {
1197 # with '};' inside yt initial data (see [1])
1198 # see [2] for an example with '};' inside ytInitialPlayerResponse
1199 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1200 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1201 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1202 'info_dict': {
1203 'id': 'CHqg6qOn4no',
1204 'ext': 'mp4',
1205 'title': 'Part 77 Sort a list of simple types in c#',
1206 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1207 'upload_date': '20130831',
1208 'uploader_id': 'kudvenkat',
1209 'uploader': 'kudvenkat',
1210 },
1211 'params': {
1212 'skip_download': True,
1213 },
1214 },
1215 {
1216 # another example of '};' in ytInitialData
1217 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1218 'only_matching': True,
1219 },
1220 {
1221 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1222 'only_matching': True,
1223 },
1224 {
1225 # https://github.com/ytdl-org/youtube-dl/pull/28094
1226 'url': 'OtqTfy26tG0',
1227 'info_dict': {
1228 'id': 'OtqTfy26tG0',
1229 'ext': 'mp4',
1230 'title': 'Burn Out',
1231 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1232 'upload_date': '20141120',
1233 'uploader': 'The Cinematic Orchestra - Topic',
1234 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1235 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1236 'artist': 'The Cinematic Orchestra',
1237 'track': 'Burn Out',
1238 'album': 'Every Day',
1239 'release_data': None,
1240 'release_year': None,
1241 },
1242 'params': {
1243 'skip_download': True,
1244 },
1245 },
1246 {
1247 # controversial video, only works with bpctr when authenticated with cookies
1248 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1249 'only_matching': True,
1250 },
1251 {
1252 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
1253 'url': 'cBvYw8_A0vQ',
1254 'info_dict': {
1255 'id': 'cBvYw8_A0vQ',
1256 'ext': 'mp4',
1257 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
1258 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
1259 'upload_date': '20201120',
1260 'uploader': 'Walk around Japan',
1261 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
1262 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
1263 },
1264 'params': {
1265 'skip_download': True,
1266 },
1267 },
1268 ]
1269
1270 def __init__(self, *args, **kwargs):
1271 super(YoutubeIE, self).__init__(*args, **kwargs)
1272 self._code_cache = {}
1273 self._player_cache = {}
1274
1275 def _signature_cache_id(self, example_sig):
1276 """ Return a string representation of a signature """
1277 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1278
1279 @classmethod
1280 def _extract_player_info(cls, player_url):
1281 for player_re in cls._PLAYER_INFO_RE:
1282 id_m = re.search(player_re, player_url)
1283 if id_m:
1284 break
1285 else:
1286 raise ExtractorError('Cannot identify player %r' % player_url)
1287 return id_m.group('id')
1288
1289 def _extract_signature_function(self, video_id, player_url, example_sig):
1290 player_id = self._extract_player_info(player_url)
1291
1292 # Read from filesystem cache
1293 func_id = 'js_%s_%s' % (
1294 player_id, self._signature_cache_id(example_sig))
1295 assert os.path.basename(func_id) == func_id
1296
1297 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1298 if cache_spec is not None:
1299 return lambda s: ''.join(s[i] for i in cache_spec)
1300
1301 if player_id not in self._code_cache:
1302 self._code_cache[player_id] = self._download_webpage(
1303 player_url, video_id,
1304 note='Downloading player ' + player_id,
1305 errnote='Download of %s failed' % player_url)
1306 code = self._code_cache[player_id]
1307 res = self._parse_sig_js(code)
1308
1309 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1310 cache_res = res(test_string)
1311 cache_spec = [ord(c) for c in cache_res]
1312
1313 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1314 return res
1315
1316 def _print_sig_code(self, func, example_sig):
1317 def gen_sig_code(idxs):
1318 def _genslice(start, end, step):
1319 starts = '' if start == 0 else str(start)
1320 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1321 steps = '' if step == 1 else (':%d' % step)
1322 return 's[%s%s%s]' % (starts, ends, steps)
1323
1324 step = None
1325 # Quelch pyflakes warnings - start will be set when step is set
1326 start = '(Never used)'
1327 for i, prev in zip(idxs[1:], idxs[:-1]):
1328 if step is not None:
1329 if i - prev == step:
1330 continue
1331 yield _genslice(start, prev, step)
1332 step = None
1333 continue
1334 if i - prev in [-1, 1]:
1335 step = i - prev
1336 start = prev
1337 continue
1338 else:
1339 yield 's[%d]' % prev
1340 if step is None:
1341 yield 's[%d]' % i
1342 else:
1343 yield _genslice(start, i, step)
1344
1345 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1346 cache_res = func(test_string)
1347 cache_spec = [ord(c) for c in cache_res]
1348 expr_code = ' + '.join(gen_sig_code(cache_spec))
1349 signature_id_tuple = '(%s)' % (
1350 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1351 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1352 ' return %s\n') % (signature_id_tuple, expr_code)
1353 self.to_screen('Extracted signature function:\n' + code)
1354
1355 def _parse_sig_js(self, jscode):
1356 funcname = self._search_regex(
1357 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1358 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1359 r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
1360 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
1361 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}\(a,\d+\)',
1362 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\(\s*""\s*\)',
1363 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1364 # Obsolete patterns
1365 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1366 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1367 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1368 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1369 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1370 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1371 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1372 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1373 jscode, 'Initial JS player signature function name', group='sig')
1374
1375 jsi = JSInterpreter(jscode)
1376 initial_function = jsi.extract_function(funcname)
1377 return lambda s: initial_function([s])
1378
1379 def _decrypt_signature(self, s, video_id, player_url):
1380 """Turn the encrypted s field into a working signature"""
1381
1382 if player_url is None:
1383 raise ExtractorError('Cannot decrypt signature without player_url')
1384
1385 if player_url.startswith('//'):
1386 player_url = 'https:' + player_url
1387 elif not re.match(r'https?://', player_url):
1388 player_url = compat_urlparse.urljoin(
1389 'https://www.youtube.com', player_url)
1390 try:
1391 player_id = (player_url, self._signature_cache_id(s))
1392 if player_id not in self._player_cache:
1393 func = self._extract_signature_function(
1394 video_id, player_url, s
1395 )
1396 self._player_cache[player_id] = func
1397 func = self._player_cache[player_id]
1398 if self._downloader.params.get('youtube_print_sig_code'):
1399 self._print_sig_code(func, s)
1400 return func(s)
1401 except Exception as e:
1402 tb = traceback.format_exc()
1403 raise ExtractorError(
1404 'Signature extraction failed: ' + tb, cause=e)
1405
1406 def _mark_watched(self, video_id, player_response):
1407 playback_url = url_or_none(try_get(
1408 player_response,
1409 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']))
1410 if not playback_url:
1411 return
1412 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1413 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1414
1415 # cpn generation algorithm is reverse engineered from base.js.
1416 # In fact it works even with dummy cpn.
1417 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1418 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1419
1420 qs.update({
1421 'ver': ['2'],
1422 'cpn': [cpn],
1423 })
1424 playback_url = compat_urlparse.urlunparse(
1425 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1426
1427 self._download_webpage(
1428 playback_url, video_id, 'Marking watched',
1429 'Unable to mark watched', fatal=False)
1430
1431 @staticmethod
1432 def _extract_urls(webpage):
1433 # Embedded YouTube player
1434 entries = [
1435 unescapeHTML(mobj.group('url'))
1436 for mobj in re.finditer(r'''(?x)
1437 (?:
1438 <iframe[^>]+?src=|
1439 data-video-url=|
1440 <embed[^>]+?src=|
1441 embedSWF\(?:\s*|
1442 <object[^>]+data=|
1443 new\s+SWFObject\(
1444 )
1445 (["\'])
1446 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1447 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1448 \1''', webpage)]
1449
1450 # lazyYT YouTube embed
1451 entries.extend(list(map(
1452 unescapeHTML,
1453 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1454
1455 # Wordpress "YouTube Video Importer" plugin
1456 matches = re.findall(r'''(?x)<div[^>]+
1457 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1458 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1459 entries.extend(m[-1] for m in matches)
1460
1461 return entries
1462
1463 @staticmethod
1464 def _extract_url(webpage):
1465 urls = YoutubeIE._extract_urls(webpage)
1466 return urls[0] if urls else None
1467
1468 @classmethod
1469 def extract_id(cls, url):
1470 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1471 if mobj is None:
1472 raise ExtractorError('Invalid URL: %s' % url)
1473 video_id = mobj.group(2)
1474 return video_id
1475
1476 def _extract_chapters_from_json(self, data, video_id, duration):
1477 chapters_list = try_get(
1478 data,
1479 lambda x: x['playerOverlays']
1480 ['playerOverlayRenderer']
1481 ['decoratedPlayerBarRenderer']
1482 ['decoratedPlayerBarRenderer']
1483 ['playerBar']
1484 ['chapteredPlayerBarRenderer']
1485 ['chapters'],
1486 list)
1487 if not chapters_list:
1488 return
1489
1490 def chapter_time(chapter):
1491 return float_or_none(
1492 try_get(
1493 chapter,
1494 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1495 int),
1496 scale=1000)
1497 chapters = []
1498 for next_num, chapter in enumerate(chapters_list, start=1):
1499 start_time = chapter_time(chapter)
1500 if start_time is None:
1501 continue
1502 end_time = (chapter_time(chapters_list[next_num])
1503 if next_num < len(chapters_list) else duration)
1504 if end_time is None:
1505 continue
1506 title = try_get(
1507 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1508 compat_str)
1509 chapters.append({
1510 'start_time': start_time,
1511 'end_time': end_time,
1512 'title': title,
1513 })
1514 return chapters
1515
1516 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
1517 return self._parse_json(self._search_regex(
1518 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
1519 regex), webpage, name, default='{}'), video_id, fatal=False)
1520
1521 @staticmethod
1522 def parse_time_text(time_text):
1523 """
1524 Parse the comment time text
1525 time_text is in the format 'X units ago (edited)'
1526 """
1527 time_text_split = time_text.split(' ')
1528 if len(time_text_split) >= 3:
1529 return datetime_from_str('now-%s%s' % (time_text_split[0], time_text_split[1]), precision='auto')
1530
1531 @staticmethod
1532 def _join_text_entries(runs):
1533 text = None
1534 for run in runs:
1535 if not isinstance(run, dict):
1536 continue
1537 sub_text = try_get(run, lambda x: x['text'], compat_str)
1538 if sub_text:
1539 if not text:
1540 text = sub_text
1541 continue
1542 text += sub_text
1543 return text
1544
1545 def _extract_comment(self, comment_renderer, parent=None):
1546 comment_id = comment_renderer.get('commentId')
1547 if not comment_id:
1548 return
1549 comment_text_runs = try_get(comment_renderer, lambda x: x['contentText']['runs']) or []
1550 text = self._join_text_entries(comment_text_runs) or ''
1551 comment_time_text = try_get(comment_renderer, lambda x: x['publishedTimeText']['runs']) or []
1552 time_text = self._join_text_entries(comment_time_text)
1553 timestamp = calendar.timegm(self.parse_time_text(time_text).timetuple())
1554 author = try_get(comment_renderer, lambda x: x['authorText']['simpleText'], compat_str)
1555 author_id = try_get(comment_renderer,
1556 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
1557 votes = str_to_int(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
1558 lambda x: x['likeCount']), compat_str)) or 0
1559 author_thumbnail = try_get(comment_renderer,
1560 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
1561
1562 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
1563 is_liked = try_get(comment_renderer, lambda x: x['isLiked'], bool)
1564 return {
1565 'id': comment_id,
1566 'text': text,
1567 'timestamp': timestamp,
1568 'time_text': time_text,
1569 'like_count': votes,
1570 'is_favorited': is_liked,
1571 'author': author,
1572 'author_id': author_id,
1573 'author_thumbnail': author_thumbnail,
1574 'author_is_uploader': author_is_uploader,
1575 'parent': parent or 'root'
1576 }
1577
1578 def _comment_entries(self, root_continuation_data, identity_token, account_syncid,
1579 session_token_list, parent=None, comment_counts=None):
1580
1581 def extract_thread(parent_renderer):
1582 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
1583 if not parent:
1584 comment_counts[2] = 0
1585 for content in contents:
1586 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
1587 comment_renderer = try_get(
1588 comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
1589 content, (lambda x: x['commentRenderer'], dict))
1590
1591 if not comment_renderer:
1592 continue
1593 comment = self._extract_comment(comment_renderer, parent)
1594 if not comment:
1595 continue
1596 comment_counts[0] += 1
1597 yield comment
1598 # Attempt to get the replies
1599 comment_replies_renderer = try_get(
1600 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
1601
1602 if comment_replies_renderer:
1603 comment_counts[2] += 1
1604 comment_entries_iter = self._comment_entries(
1605 comment_replies_renderer, identity_token, account_syncid,
1606 parent=comment.get('id'), session_token_list=session_token_list,
1607 comment_counts=comment_counts)
1608
1609 for reply_comment in comment_entries_iter:
1610 yield reply_comment
1611
1612 if not comment_counts:
1613 # comment so far, est. total comments, current comment thread #
1614 comment_counts = [0, 0, 0]
1615 headers = self._DEFAULT_BASIC_API_HEADERS.copy()
1616
1617 # TODO: Generalize the download code with TabIE
1618 if identity_token:
1619 headers['x-youtube-identity-token'] = identity_token
1620
1621 if account_syncid:
1622 headers['X-Goog-PageId'] = account_syncid
1623 headers['X-Goog-AuthUser'] = 0
1624
1625 continuation = YoutubeTabIE._extract_continuation(root_continuation_data) # TODO
1626 first_continuation = False
1627 if parent is None:
1628 first_continuation = True
1629
1630 for page_num in itertools.count(0):
1631 if not continuation:
1632 break
1633 retries = self._downloader.params.get('extractor_retries', 3)
1634 count = -1
1635 last_error = None
1636
1637 while count < retries:
1638 count += 1
1639 if last_error:
1640 self.report_warning('%s. Retrying ...' % last_error)
1641 try:
1642 query = {
1643 'ctoken': continuation['ctoken'],
1644 'pbj': 1,
1645 'type': 'next',
1646 }
1647 if parent:
1648 query['action_get_comment_replies'] = 1
1649 else:
1650 query['action_get_comments'] = 1
1651
1652 comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
1653 if page_num == 0:
1654 if first_continuation:
1655 note_prefix = 'Downloading initial comment continuation page'
1656 else:
1657 note_prefix = ' Downloading comment reply thread %d %s' % (comment_counts[2], comment_prog_str)
1658 else:
1659 note_prefix = '%sDownloading comment%s page %d %s' % (
1660 ' ' if parent else '',
1661 ' replies' if parent else '',
1662 page_num,
1663 comment_prog_str)
1664
1665 browse = self._download_json(
1666 'https://www.youtube.com/comment_service_ajax', None,
1667 '%s %s' % (note_prefix, '(retry #%d)' % count if count else ''),
1668 headers=headers, query=query,
1669 data=urlencode_postdata({
1670 'session_token': session_token_list[0]
1671 }))
1672 except ExtractorError as e:
1673 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404, 413):
1674 if e.cause.code == 413:
1675 self.report_warning('Assumed end of comments (received HTTP Error 413)')
1676 return
1677 # Downloading page may result in intermittent 5xx HTTP error
1678 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
1679 last_error = 'HTTP Error %s' % e.cause.code
1680 if e.cause.code == 404:
1681 last_error = last_error + ' (this API is probably deprecated)'
1682 if count < retries:
1683 continue
1684 raise
1685 else:
1686 session_token = try_get(browse, lambda x: x['xsrf_token'], compat_str)
1687 if session_token:
1688 session_token_list[0] = session_token
1689
1690 response = try_get(browse,
1691 (lambda x: x['response'],
1692 lambda x: x[1]['response'])) or {}
1693
1694 if response.get('continuationContents'):
1695 break
1696
1697 # YouTube sometimes gives reload: now json if something went wrong (e.g. bad auth)
1698 if browse.get('reload'):
1699 raise ExtractorError('Invalid or missing params in continuation request', expected=False)
1700
1701 # TODO: not tested, merged from old extractor
1702 err_msg = browse.get('externalErrorMessage')
1703 if err_msg:
1704 raise ExtractorError('YouTube said: %s' % err_msg, expected=False)
1705
1706 # Youtube sometimes sends incomplete data
1707 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1708 last_error = 'Incomplete data received'
1709 if count >= retries:
1710 self._downloader.report_error(last_error)
1711
1712 if not response:
1713 break
1714
1715 known_continuation_renderers = {
1716 'itemSectionContinuation': extract_thread,
1717 'commentRepliesContinuation': extract_thread
1718 }
1719
1720 # extract next root continuation from the results
1721 continuation_contents = try_get(
1722 response, lambda x: x['continuationContents'], dict) or {}
1723
1724 for key, value in continuation_contents.items():
1725 if key not in known_continuation_renderers:
1726 continue
1727 continuation_renderer = value
1728
1729 if first_continuation:
1730 first_continuation = False
1731 expected_comment_count = try_get(
1732 continuation_renderer,
1733 (lambda x: x['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'],
1734 lambda x: x['header']['commentsHeaderRenderer']['commentsCount']['runs'][0]['text']),
1735 compat_str)
1736
1737 if expected_comment_count:
1738 comment_counts[1] = str_to_int(expected_comment_count)
1739 self.to_screen('Downloading ~%d comments' % str_to_int(expected_comment_count))
1740 yield comment_counts[1]
1741
1742 # TODO: cli arg.
1743 # 1/True for newest, 0/False for popular (default)
1744 comment_sort_index = int(True)
1745 sort_continuation_renderer = try_get(
1746 continuation_renderer,
1747 lambda x: x['header']['commentsHeaderRenderer']['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems']
1748 [comment_sort_index]['continuation']['reloadContinuationData'], dict)
1749 # If this fails, the initial continuation page
1750 # starts off with popular anyways.
1751 if sort_continuation_renderer:
1752 continuation = YoutubeTabIE._build_continuation_query(
1753 continuation=sort_continuation_renderer.get('continuation'),
1754 ctp=sort_continuation_renderer.get('clickTrackingParams'))
1755 self.to_screen('Sorting comments by %s' % ('popular' if comment_sort_index == 0 else 'newest'))
1756 break
1757
1758 for entry in known_continuation_renderers[key](continuation_renderer):
1759 yield entry
1760
1761 continuation = YoutubeTabIE._extract_continuation(continuation_renderer) # TODO
1762 break
1763
1764 def _extract_comments(self, ytcfg, video_id, contents, webpage, xsrf_token):
1765 """Entry for comment extraction"""
1766 comments = []
1767 known_entry_comment_renderers = (
1768 'itemSectionRenderer',
1769 )
1770 estimated_total = 0
1771 for entry in contents:
1772 for key, renderer in entry.items():
1773 if key not in known_entry_comment_renderers:
1774 continue
1775
1776 comment_iter = self._comment_entries(
1777 renderer,
1778 identity_token=self._extract_identity_token(webpage, item_id=video_id),
1779 account_syncid=self._extract_account_syncid(ytcfg),
1780 session_token_list=[xsrf_token])
1781
1782 for comment in comment_iter:
1783 if isinstance(comment, int):
1784 estimated_total = comment
1785 continue
1786 comments.append(comment)
1787 break
1788 self.to_screen('Downloaded %d/%d comments' % (len(comments), estimated_total))
1789 return {
1790 'comments': comments,
1791 'comment_count': len(comments),
1792 }
1793
1794 def _real_extract(self, url):
1795 url, smuggled_data = unsmuggle_url(url, {})
1796 video_id = self._match_id(url)
1797 base_url = self.http_scheme() + '//www.youtube.com/'
1798 webpage_url = base_url + 'watch?v=' + video_id
1799 webpage = self._download_webpage(
1800 webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False)
1801
1802 player_response = None
1803 if webpage:
1804 player_response = self._extract_yt_initial_variable(
1805 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
1806 video_id, 'initial player response')
1807 if not player_response:
1808 player_response = self._call_api(
1809 'player', {'videoId': video_id}, video_id)
1810
1811 playability_status = player_response.get('playabilityStatus') or {}
1812 if playability_status.get('reason') == 'Sign in to confirm your age':
1813 pr = self._parse_json(try_get(compat_parse_qs(
1814 self._download_webpage(
1815 base_url + 'get_video_info', video_id,
1816 'Refetching age-gated info webpage',
1817 'unable to download video info webpage', query={
1818 'video_id': video_id,
1819 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1820 }, fatal=False)),
1821 lambda x: x['player_response'][0],
1822 compat_str) or '{}', video_id)
1823 if pr:
1824 player_response = pr
1825
1826 trailer_video_id = try_get(
1827 playability_status,
1828 lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'],
1829 compat_str)
1830 if trailer_video_id:
1831 return self.url_result(
1832 trailer_video_id, self.ie_key(), trailer_video_id)
1833
1834 def get_text(x):
1835 if not x:
1836 return
1837 text = x.get('simpleText')
1838 if text and isinstance(text, compat_str):
1839 return text
1840 runs = x.get('runs')
1841 if not isinstance(runs, list):
1842 return
1843 return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
1844
1845 search_meta = (
1846 lambda x: self._html_search_meta(x, webpage, default=None)) \
1847 if webpage else lambda x: None
1848
1849 video_details = player_response.get('videoDetails') or {}
1850 microformat = try_get(
1851 player_response,
1852 lambda x: x['microformat']['playerMicroformatRenderer'],
1853 dict) or {}
1854 video_title = video_details.get('title') \
1855 or get_text(microformat.get('title')) \
1856 or search_meta(['og:title', 'twitter:title', 'title'])
1857 video_description = video_details.get('shortDescription')
1858
1859 if not smuggled_data.get('force_singlefeed', False):
1860 if not self._downloader.params.get('noplaylist'):
1861 multifeed_metadata_list = try_get(
1862 player_response,
1863 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1864 compat_str)
1865 if multifeed_metadata_list:
1866 entries = []
1867 feed_ids = []
1868 for feed in multifeed_metadata_list.split(','):
1869 # Unquote should take place before split on comma (,) since textual
1870 # fields may contain comma as well (see
1871 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1872 feed_data = compat_parse_qs(
1873 compat_urllib_parse_unquote_plus(feed))
1874
1875 def feed_entry(name):
1876 return try_get(
1877 feed_data, lambda x: x[name][0], compat_str)
1878
1879 feed_id = feed_entry('id')
1880 if not feed_id:
1881 continue
1882 feed_title = feed_entry('title')
1883 title = video_title
1884 if feed_title:
1885 title += ' (%s)' % feed_title
1886 entries.append({
1887 '_type': 'url_transparent',
1888 'ie_key': 'Youtube',
1889 'url': smuggle_url(
1890 base_url + 'watch?v=' + feed_data['id'][0],
1891 {'force_singlefeed': True}),
1892 'title': title,
1893 })
1894 feed_ids.append(feed_id)
1895 self.to_screen(
1896 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1897 % (', '.join(feed_ids), video_id))
1898 return self.playlist_result(
1899 entries, video_id, video_title, video_description)
1900 else:
1901 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1902
1903 formats = []
1904 itags = []
1905 itag_qualities = {}
1906 player_url = None
1907 q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'])
1908 streaming_data = player_response.get('streamingData') or {}
1909 streaming_formats = streaming_data.get('formats') or []
1910 streaming_formats.extend(streaming_data.get('adaptiveFormats') or [])
1911 for fmt in streaming_formats:
1912 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
1913 continue
1914
1915 itag = str_or_none(fmt.get('itag'))
1916 quality = fmt.get('quality')
1917 if itag and quality:
1918 itag_qualities[itag] = quality
1919 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
1920 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
1921 # number of fragment that would subsequently requested with (`&sq=N`)
1922 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
1923 continue
1924
1925 fmt_url = fmt.get('url')
1926 if not fmt_url:
1927 sc = compat_parse_qs(fmt.get('signatureCipher'))
1928 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
1929 encrypted_sig = try_get(sc, lambda x: x['s'][0])
1930 if not (sc and fmt_url and encrypted_sig):
1931 continue
1932 if not player_url:
1933 if not webpage:
1934 continue
1935 player_url = self._search_regex(
1936 r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"',
1937 webpage, 'player URL', fatal=False)
1938 if not player_url:
1939 continue
1940 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
1941 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
1942 fmt_url += '&' + sp + '=' + signature
1943
1944 if itag:
1945 itags.append(itag)
1946 tbr = float_or_none(
1947 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
1948 dct = {
1949 'asr': int_or_none(fmt.get('audioSampleRate')),
1950 'filesize': int_or_none(fmt.get('contentLength')),
1951 'format_id': itag,
1952 'format_note': fmt.get('qualityLabel') or quality,
1953 'fps': int_or_none(fmt.get('fps')),
1954 'height': int_or_none(fmt.get('height')),
1955 'quality': q(quality),
1956 'tbr': tbr,
1957 'url': fmt_url,
1958 'width': fmt.get('width'),
1959 }
1960 mimetype = fmt.get('mimeType')
1961 if mimetype:
1962 mobj = re.match(
1963 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype)
1964 if mobj:
1965 dct['ext'] = mimetype2ext(mobj.group(1))
1966 dct.update(parse_codecs(mobj.group(2)))
1967 no_audio = dct.get('acodec') == 'none'
1968 no_video = dct.get('vcodec') == 'none'
1969 if no_audio:
1970 dct['vbr'] = tbr
1971 if no_video:
1972 dct['abr'] = tbr
1973 if no_audio or no_video:
1974 dct['downloader_options'] = {
1975 # Youtube throttles chunks >~10M
1976 'http_chunk_size': 10485760,
1977 }
1978 if dct.get('ext'):
1979 dct['container'] = dct['ext'] + '_dash'
1980 formats.append(dct)
1981
1982 hls_manifest_url = streaming_data.get('hlsManifestUrl')
1983 if hls_manifest_url:
1984 for f in self._extract_m3u8_formats(
1985 hls_manifest_url, video_id, 'mp4', fatal=False):
1986 itag = self._search_regex(
1987 r'/itag/(\d+)', f['url'], 'itag', default=None)
1988 if itag:
1989 f['format_id'] = itag
1990 formats.append(f)
1991
1992 if self._downloader.params.get('youtube_include_dash_manifest', True):
1993 dash_manifest_url = streaming_data.get('dashManifestUrl')
1994 if dash_manifest_url:
1995 for f in self._extract_mpd_formats(
1996 dash_manifest_url, video_id, fatal=False):
1997 itag = f['format_id']
1998 if itag in itags:
1999 continue
2000 if itag in itag_qualities:
2001 # Not actually usefull since the sorting is already done with "quality,res,fps,codec"
2002 # but kept to maintain feature parity (and code similarity) with youtube-dl
2003 # Remove if this causes any issues with sorting in future
2004 f['quality'] = q(itag_qualities[itag])
2005 filesize = int_or_none(self._search_regex(
2006 r'/clen/(\d+)', f.get('fragment_base_url')
2007 or f['url'], 'file size', default=None))
2008 if filesize:
2009 f['filesize'] = filesize
2010 formats.append(f)
2011
2012 if not formats:
2013 if not self._downloader.params.get('allow_unplayable_formats') and streaming_data.get('licenseInfos'):
2014 raise ExtractorError(
2015 'This video is DRM protected.', expected=True)
2016 pemr = try_get(
2017 playability_status,
2018 lambda x: x['errorScreen']['playerErrorMessageRenderer'],
2019 dict) or {}
2020 reason = get_text(pemr.get('reason')) or playability_status.get('reason')
2021 subreason = pemr.get('subreason')
2022 if subreason:
2023 subreason = clean_html(get_text(subreason))
2024 if subreason == 'The uploader has not made this video available in your country.':
2025 countries = microformat.get('availableCountries')
2026 if not countries:
2027 regions_allowed = search_meta('regionsAllowed')
2028 countries = regions_allowed.split(',') if regions_allowed else None
2029 self.raise_geo_restricted(
2030 subreason, countries)
2031 reason += '\n' + subreason
2032 if reason:
2033 raise ExtractorError(reason, expected=True)
2034
2035 self._sort_formats(formats)
2036
2037 keywords = video_details.get('keywords') or []
2038 if not keywords and webpage:
2039 keywords = [
2040 unescapeHTML(m.group('content'))
2041 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
2042 for keyword in keywords:
2043 if keyword.startswith('yt:stretch='):
2044 w, h = keyword.split('=')[1].split(':')
2045 w, h = int(w), int(h)
2046 if w > 0 and h > 0:
2047 ratio = w / h
2048 for f in formats:
2049 if f.get('vcodec') != 'none':
2050 f['stretched_ratio'] = ratio
2051
2052 thumbnails = []
2053 for container in (video_details, microformat):
2054 for thumbnail in (try_get(
2055 container,
2056 lambda x: x['thumbnail']['thumbnails'], list) or []):
2057 thumbnail_url = thumbnail.get('url')
2058 if not thumbnail_url:
2059 continue
2060 thumbnails.append({
2061 'height': int_or_none(thumbnail.get('height')),
2062 'url': thumbnail_url,
2063 'width': int_or_none(thumbnail.get('width')),
2064 })
2065 if thumbnails:
2066 break
2067 else:
2068 thumbnail = search_meta(['og:image', 'twitter:image'])
2069 if thumbnail:
2070 thumbnails = [{'url': thumbnail}]
2071
2072 category = microformat.get('category') or search_meta('genre')
2073 channel_id = video_details.get('channelId') \
2074 or microformat.get('externalChannelId') \
2075 or search_meta('channelId')
2076 duration = int_or_none(
2077 video_details.get('lengthSeconds')
2078 or microformat.get('lengthSeconds')) \
2079 or parse_duration(search_meta('duration'))
2080 is_live = video_details.get('isLive')
2081 owner_profile_url = microformat.get('ownerProfileUrl')
2082
2083 info = {
2084 'id': video_id,
2085 'title': self._live_title(video_title) if is_live else video_title,
2086 'formats': formats,
2087 'thumbnails': thumbnails,
2088 'description': video_description,
2089 'upload_date': unified_strdate(
2090 microformat.get('uploadDate')
2091 or search_meta('uploadDate')),
2092 'uploader': video_details['author'],
2093 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
2094 'uploader_url': owner_profile_url,
2095 'channel_id': channel_id,
2096 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None,
2097 'duration': duration,
2098 'view_count': int_or_none(
2099 video_details.get('viewCount')
2100 or microformat.get('viewCount')
2101 or search_meta('interactionCount')),
2102 'average_rating': float_or_none(video_details.get('averageRating')),
2103 'age_limit': 18 if (
2104 microformat.get('isFamilySafe') is False
2105 or search_meta('isFamilyFriendly') == 'false'
2106 or search_meta('og:restrictions:age') == '18+') else 0,
2107 'webpage_url': webpage_url,
2108 'categories': [category] if category else None,
2109 'tags': keywords,
2110 'is_live': is_live,
2111 'playable_in_embed': playability_status.get('playableInEmbed'),
2112 'was_live': video_details.get('isLiveContent'),
2113 }
2114
2115 pctr = try_get(
2116 player_response,
2117 lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
2118 subtitles = {}
2119 if pctr:
2120 def process_language(container, base_url, lang_code, query):
2121 lang_subs = []
2122 for fmt in self._SUBTITLE_FORMATS:
2123 query.update({
2124 'fmt': fmt,
2125 })
2126 lang_subs.append({
2127 'ext': fmt,
2128 'url': update_url_query(base_url, query),
2129 })
2130 container[lang_code] = lang_subs
2131
2132 for caption_track in (pctr.get('captionTracks') or []):
2133 base_url = caption_track.get('baseUrl')
2134 if not base_url:
2135 continue
2136 if caption_track.get('kind') != 'asr':
2137 lang_code = caption_track.get('languageCode')
2138 if not lang_code:
2139 continue
2140 process_language(
2141 subtitles, base_url, lang_code, {})
2142 continue
2143 automatic_captions = {}
2144 for translation_language in (pctr.get('translationLanguages') or []):
2145 translation_language_code = translation_language.get('languageCode')
2146 if not translation_language_code:
2147 continue
2148 process_language(
2149 automatic_captions, base_url, translation_language_code,
2150 {'tlang': translation_language_code})
2151 info['automatic_captions'] = automatic_captions
2152 info['subtitles'] = subtitles
2153
2154 parsed_url = compat_urllib_parse_urlparse(url)
2155 for component in [parsed_url.fragment, parsed_url.query]:
2156 query = compat_parse_qs(component)
2157 for k, v in query.items():
2158 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
2159 d_k += '_time'
2160 if d_k not in info and k in s_ks:
2161 info[d_k] = parse_duration(query[k][0])
2162
2163 # Youtube Music Auto-generated description
2164 if video_description:
2165 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)
2166 if mobj:
2167 release_year = mobj.group('release_year')
2168 release_date = mobj.group('release_date')
2169 if release_date:
2170 release_date = release_date.replace('-', '')
2171 if not release_year:
2172 release_year = release_date[:4]
2173 info.update({
2174 'album': mobj.group('album'.strip()),
2175 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
2176 'track': mobj.group('track').strip(),
2177 'release_date': release_date,
2178 'release_year': int_or_none(release_year),
2179 })
2180
2181 initial_data = None
2182 if webpage:
2183 initial_data = self._extract_yt_initial_variable(
2184 webpage, self._YT_INITIAL_DATA_RE, video_id,
2185 'yt initial data')
2186 if not initial_data:
2187 initial_data = self._call_api(
2188 'next', {'videoId': video_id}, video_id, fatal=False)
2189
2190 if not is_live:
2191 try:
2192 # This will error if there is no livechat
2193 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2194 info['subtitles']['live_chat'] = [{
2195 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
2196 'video_id': video_id,
2197 'ext': 'json',
2198 'protocol': 'youtube_live_chat_replay',
2199 }]
2200 except (KeyError, IndexError, TypeError):
2201 pass
2202
2203 if initial_data:
2204 chapters = self._extract_chapters_from_json(
2205 initial_data, video_id, duration)
2206 if not chapters:
2207 for engagment_pannel in (initial_data.get('engagementPanels') or []):
2208 contents = try_get(
2209 engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'],
2210 list)
2211 if not contents:
2212 continue
2213
2214 def chapter_time(mmlir):
2215 return parse_duration(
2216 get_text(mmlir.get('timeDescription')))
2217
2218 chapters = []
2219 for next_num, content in enumerate(contents, start=1):
2220 mmlir = content.get('macroMarkersListItemRenderer') or {}
2221 start_time = chapter_time(mmlir)
2222 end_time = chapter_time(try_get(
2223 contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \
2224 if next_num < len(contents) else duration
2225 if start_time is None or end_time is None:
2226 continue
2227 chapters.append({
2228 'start_time': start_time,
2229 'end_time': end_time,
2230 'title': get_text(mmlir.get('title')),
2231 })
2232 if chapters:
2233 break
2234 if chapters:
2235 info['chapters'] = chapters
2236
2237 contents = try_get(
2238 initial_data,
2239 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
2240 list) or []
2241 for content in contents:
2242 vpir = content.get('videoPrimaryInfoRenderer')
2243 if vpir:
2244 stl = vpir.get('superTitleLink')
2245 if stl:
2246 stl = get_text(stl)
2247 if try_get(
2248 vpir,
2249 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
2250 info['location'] = stl
2251 else:
2252 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
2253 if mobj:
2254 info.update({
2255 'series': mobj.group(1),
2256 'season_number': int(mobj.group(2)),
2257 'episode_number': int(mobj.group(3)),
2258 })
2259 for tlb in (try_get(
2260 vpir,
2261 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
2262 list) or []):
2263 tbr = tlb.get('toggleButtonRenderer') or {}
2264 for getter, regex in [(
2265 lambda x: x['defaultText']['accessibility']['accessibilityData'],
2266 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
2267 lambda x: x['accessibility'],
2268 lambda x: x['accessibilityData']['accessibilityData'],
2269 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
2270 label = (try_get(tbr, getter, dict) or {}).get('label')
2271 if label:
2272 mobj = re.match(regex, label)
2273 if mobj:
2274 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
2275 break
2276 sbr_tooltip = try_get(
2277 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
2278 if sbr_tooltip:
2279 like_count, dislike_count = sbr_tooltip.split(' / ')
2280 info.update({
2281 'like_count': str_to_int(like_count),
2282 'dislike_count': str_to_int(dislike_count),
2283 })
2284 vsir = content.get('videoSecondaryInfoRenderer')
2285 if vsir:
2286 info['channel'] = get_text(try_get(
2287 vsir,
2288 lambda x: x['owner']['videoOwnerRenderer']['title'],
2289 dict))
2290 rows = try_get(
2291 vsir,
2292 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
2293 list) or []
2294 multiple_songs = False
2295 for row in rows:
2296 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2297 multiple_songs = True
2298 break
2299 for row in rows:
2300 mrr = row.get('metadataRowRenderer') or {}
2301 mrr_title = mrr.get('title')
2302 if not mrr_title:
2303 continue
2304 mrr_title = get_text(mrr['title'])
2305 mrr_contents_text = get_text(mrr['contents'][0])
2306 if mrr_title == 'License':
2307 info['license'] = mrr_contents_text
2308 elif not multiple_songs:
2309 if mrr_title == 'Album':
2310 info['album'] = mrr_contents_text
2311 elif mrr_title == 'Artist':
2312 info['artist'] = mrr_contents_text
2313 elif mrr_title == 'Song':
2314 info['track'] = mrr_contents_text
2315
2316 fallbacks = {
2317 'channel': 'uploader',
2318 'channel_id': 'uploader_id',
2319 'channel_url': 'uploader_url',
2320 }
2321 for to, frm in fallbacks.items():
2322 if not info.get(to):
2323 info[to] = info.get(frm)
2324
2325 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
2326 v = info.get(s_k)
2327 if v:
2328 info[d_k] = v
2329
2330 is_private = bool_or_none(video_details.get('isPrivate'))
2331 is_unlisted = bool_or_none(microformat.get('isUnlisted'))
2332 is_membersonly = None
2333 is_premium = None
2334 if initial_data and is_private is not None:
2335 is_membersonly = False
2336 is_premium = False
2337 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list)
2338 for content in contents or []:
2339 badges = try_get(content, lambda x: x['videoPrimaryInfoRenderer']['badges'], list)
2340 for badge in badges or []:
2341 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label']) or ''
2342 if label.lower() == 'members only':
2343 is_membersonly = True
2344 break
2345 elif label.lower() == 'premium':
2346 is_premium = True
2347 break
2348 if is_membersonly or is_premium:
2349 break
2350
2351 # TODO: Add this for playlists
2352 info['availability'] = self._availability(
2353 is_private=is_private,
2354 needs_premium=is_premium,
2355 needs_subscription=is_membersonly,
2356 needs_auth=info['age_limit'] >= 18,
2357 is_unlisted=None if is_private is None else is_unlisted)
2358
2359 # get xsrf for annotations or comments
2360 get_annotations = self._downloader.params.get('writeannotations', False)
2361 get_comments = self._downloader.params.get('getcomments', False)
2362 if get_annotations or get_comments:
2363 xsrf_token = None
2364 ytcfg = self._extract_ytcfg(video_id, webpage)
2365 if ytcfg:
2366 xsrf_token = try_get(ytcfg, lambda x: x['XSRF_TOKEN'], compat_str)
2367 if not xsrf_token:
2368 xsrf_token = self._search_regex(
2369 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>(?:(?!\2).)+)\2',
2370 webpage, 'xsrf token', group='xsrf_token', fatal=False)
2371
2372 # annotations
2373 if get_annotations:
2374 invideo_url = try_get(
2375 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2376 if xsrf_token and invideo_url:
2377 xsrf_field_name = None
2378 if ytcfg:
2379 xsrf_field_name = try_get(ytcfg, lambda x: x['XSRF_FIELD_NAME'], compat_str)
2380 if not xsrf_field_name:
2381 xsrf_field_name = self._search_regex(
2382 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2383 webpage, 'xsrf field name',
2384 group='xsrf_field_name', default='session_token')
2385 info['annotations'] = self._download_webpage(
2386 self._proto_relative_url(invideo_url),
2387 video_id, note='Downloading annotations',
2388 errnote='Unable to download video annotations', fatal=False,
2389 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2390
2391 if get_comments:
2392 info['__post_extractor'] = lambda: self._extract_comments(ytcfg, video_id, contents, webpage, xsrf_token)
2393
2394 self.mark_watched(video_id, player_response)
2395
2396 return info
2397
2398
2399 class YoutubeTabIE(YoutubeBaseInfoExtractor):
2400 IE_DESC = 'YouTube.com tab'
2401 _VALID_URL = r'''(?x)
2402 https?://
2403 (?:\w+\.)?
2404 (?:
2405 youtube(?:kids)?\.com|
2406 invidio\.us
2407 )/
2408 (?:
2409 (?:channel|c|user)/|
2410 (?P<not_channel>
2411 feed/|hashtag/|
2412 (?:playlist|watch)\?.*?\blist=
2413 )|
2414 (?!(?:%s)\b) # Direct URLs
2415 )
2416 (?P<id>[^/?\#&]+)
2417 ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
2418 IE_NAME = 'youtube:tab'
2419
2420 _TESTS = [{
2421 # playlists, multipage
2422 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
2423 'playlist_mincount': 94,
2424 'info_dict': {
2425 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2426 'title': 'Игорь Клейнер - Playlists',
2427 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2428 'uploader': 'Игорь Клейнер',
2429 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2430 },
2431 }, {
2432 # playlists, multipage, different order
2433 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2434 'playlist_mincount': 94,
2435 'info_dict': {
2436 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2437 'title': 'Игорь Клейнер - Playlists',
2438 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2439 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2440 'uploader': 'Игорь Клейнер',
2441 },
2442 }, {
2443 # playlists, singlepage
2444 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2445 'playlist_mincount': 4,
2446 'info_dict': {
2447 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2448 'title': 'ThirstForScience - Playlists',
2449 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
2450 'uploader': 'ThirstForScience',
2451 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2452 }
2453 }, {
2454 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
2455 'only_matching': True,
2456 }, {
2457 # basic, single video playlist
2458 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2459 'info_dict': {
2460 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2461 'uploader': 'Sergey M.',
2462 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2463 'title': 'youtube-dl public playlist',
2464 },
2465 'playlist_count': 1,
2466 }, {
2467 # empty playlist
2468 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2469 'info_dict': {
2470 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2471 'uploader': 'Sergey M.',
2472 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2473 'title': 'youtube-dl empty playlist',
2474 },
2475 'playlist_count': 0,
2476 }, {
2477 # Home tab
2478 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
2479 'info_dict': {
2480 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2481 'title': 'lex will - Home',
2482 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2483 'uploader': 'lex will',
2484 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2485 },
2486 'playlist_mincount': 2,
2487 }, {
2488 # Videos tab
2489 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
2490 'info_dict': {
2491 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2492 'title': 'lex will - Videos',
2493 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2494 'uploader': 'lex will',
2495 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2496 },
2497 'playlist_mincount': 975,
2498 }, {
2499 # Videos tab, sorted by popular
2500 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
2501 'info_dict': {
2502 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2503 'title': 'lex will - Videos',
2504 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2505 'uploader': 'lex will',
2506 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2507 },
2508 'playlist_mincount': 199,
2509 }, {
2510 # Playlists tab
2511 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
2512 'info_dict': {
2513 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2514 'title': 'lex will - Playlists',
2515 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2516 'uploader': 'lex will',
2517 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2518 },
2519 'playlist_mincount': 17,
2520 }, {
2521 # Community tab
2522 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
2523 'info_dict': {
2524 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2525 'title': 'lex will - Community',
2526 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2527 'uploader': 'lex will',
2528 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2529 },
2530 'playlist_mincount': 18,
2531 }, {
2532 # Channels tab
2533 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
2534 'info_dict': {
2535 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2536 'title': 'lex will - Channels',
2537 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2538 'uploader': 'lex will',
2539 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2540 },
2541 'playlist_mincount': 12,
2542 }, {
2543 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2544 'only_matching': True,
2545 }, {
2546 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2547 'only_matching': True,
2548 }, {
2549 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2550 'only_matching': True,
2551 }, {
2552 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2553 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2554 'info_dict': {
2555 'title': '29C3: Not my department',
2556 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2557 'uploader': 'Christiaan008',
2558 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
2559 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
2560 },
2561 'playlist_count': 96,
2562 }, {
2563 'note': 'Large playlist',
2564 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2565 'info_dict': {
2566 'title': 'Uploads from Cauchemar',
2567 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2568 'uploader': 'Cauchemar',
2569 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
2570 },
2571 'playlist_mincount': 1123,
2572 }, {
2573 # even larger playlist, 8832 videos
2574 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
2575 'only_matching': True,
2576 }, {
2577 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2578 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2579 'info_dict': {
2580 'title': 'Uploads from Interstellar Movie',
2581 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2582 'uploader': 'Interstellar Movie',
2583 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
2584 },
2585 'playlist_mincount': 21,
2586 }, {
2587 # https://github.com/ytdl-org/youtube-dl/issues/21844
2588 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2589 'info_dict': {
2590 'title': 'Data Analysis with Dr Mike Pound',
2591 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2592 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
2593 'uploader': 'Computerphile',
2594 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
2595 },
2596 'playlist_mincount': 11,
2597 }, {
2598 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2599 'only_matching': True,
2600 }, {
2601 # Playlist URL that does not actually serve a playlist
2602 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2603 'info_dict': {
2604 'id': 'FqZTN594JQw',
2605 'ext': 'webm',
2606 'title': "Smiley's People 01 detective, Adventure Series, Action",
2607 'uploader': 'STREEM',
2608 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2609 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2610 'upload_date': '20150526',
2611 'license': 'Standard YouTube License',
2612 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2613 'categories': ['People & Blogs'],
2614 'tags': list,
2615 'view_count': int,
2616 'like_count': int,
2617 'dislike_count': int,
2618 },
2619 'params': {
2620 'skip_download': True,
2621 },
2622 'skip': 'This video is not available.',
2623 'add_ie': [YoutubeIE.ie_key()],
2624 }, {
2625 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2626 'only_matching': True,
2627 }, {
2628 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
2629 'only_matching': True,
2630 }, {
2631 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
2632 'info_dict': {
2633 'id': '9Auq9mYxFEE',
2634 'ext': 'mp4',
2635 'title': compat_str,
2636 'uploader': 'Sky News',
2637 'uploader_id': 'skynews',
2638 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
2639 'upload_date': '20191102',
2640 'description': 'md5:85ddd75d888674631aaf9599a9a0b0ae',
2641 'categories': ['News & Politics'],
2642 'tags': list,
2643 'like_count': int,
2644 'dislike_count': int,
2645 },
2646 'params': {
2647 'skip_download': True,
2648 },
2649 }, {
2650 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2651 'info_dict': {
2652 'id': 'a48o2S1cPoo',
2653 'ext': 'mp4',
2654 'title': 'The Young Turks - Live Main Show',
2655 'uploader': 'The Young Turks',
2656 'uploader_id': 'TheYoungTurks',
2657 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2658 'upload_date': '20150715',
2659 'license': 'Standard YouTube License',
2660 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2661 'categories': ['News & Politics'],
2662 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2663 'like_count': int,
2664 'dislike_count': int,
2665 },
2666 'params': {
2667 'skip_download': True,
2668 },
2669 'only_matching': True,
2670 }, {
2671 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2672 'only_matching': True,
2673 }, {
2674 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2675 'only_matching': True,
2676 }, {
2677 'url': 'https://www.youtube.com/feed/trending',
2678 'only_matching': True,
2679 }, {
2680 # needs auth
2681 'url': 'https://www.youtube.com/feed/library',
2682 'only_matching': True,
2683 }, {
2684 # needs auth
2685 'url': 'https://www.youtube.com/feed/history',
2686 'only_matching': True,
2687 }, {
2688 # needs auth
2689 'url': 'https://www.youtube.com/feed/subscriptions',
2690 'only_matching': True,
2691 }, {
2692 # needs auth
2693 'url': 'https://www.youtube.com/feed/watch_later',
2694 'only_matching': True,
2695 }, {
2696 # no longer available?
2697 'url': 'https://www.youtube.com/feed/recommended',
2698 'only_matching': True,
2699 }, {
2700 # inline playlist with not always working continuations
2701 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
2702 'only_matching': True,
2703 }, {
2704 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8',
2705 'only_matching': True,
2706 }, {
2707 'url': 'https://www.youtube.com/course',
2708 'only_matching': True,
2709 }, {
2710 'url': 'https://www.youtube.com/zsecurity',
2711 'only_matching': True,
2712 }, {
2713 'url': 'http://www.youtube.com/NASAgovVideo/videos',
2714 'only_matching': True,
2715 }, {
2716 'url': 'https://www.youtube.com/TheYoungTurks/live',
2717 'only_matching': True,
2718 }, {
2719 'url': 'https://www.youtube.com/hashtag/cctv9',
2720 'info_dict': {
2721 'id': 'cctv9',
2722 'title': '#cctv9',
2723 },
2724 'playlist_mincount': 350,
2725 }]
2726
2727 @classmethod
2728 def suitable(cls, url):
2729 return False if YoutubeIE.suitable(url) else super(
2730 YoutubeTabIE, cls).suitable(url)
2731
2732 def _extract_channel_id(self, webpage):
2733 channel_id = self._html_search_meta(
2734 'channelId', webpage, 'channel id', default=None)
2735 if channel_id:
2736 return channel_id
2737 channel_url = self._html_search_meta(
2738 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
2739 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
2740 'twitter:app:url:googleplay'), webpage, 'channel url')
2741 return self._search_regex(
2742 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
2743 channel_url, 'channel id')
2744
2745 @staticmethod
2746 def _extract_basic_item_renderer(item):
2747 # Modified from _extract_grid_item_renderer
2748 known_renderers = (
2749 'playlistRenderer', 'videoRenderer', 'channelRenderer',
2750 'gridPlaylistRenderer', 'gridVideoRenderer', 'gridChannelRenderer'
2751 )
2752 for key, renderer in item.items():
2753 if key not in known_renderers:
2754 continue
2755 return renderer
2756
2757 def _grid_entries(self, grid_renderer):
2758 for item in grid_renderer['items']:
2759 if not isinstance(item, dict):
2760 continue
2761 renderer = self._extract_basic_item_renderer(item)
2762 if not isinstance(renderer, dict):
2763 continue
2764 title = try_get(
2765 renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2766 # playlist
2767 playlist_id = renderer.get('playlistId')
2768 if playlist_id:
2769 yield self.url_result(
2770 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2771 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
2772 video_title=title)
2773 # video
2774 video_id = renderer.get('videoId')
2775 if video_id:
2776 yield self._extract_video(renderer)
2777 # channel
2778 channel_id = renderer.get('channelId')
2779 if channel_id:
2780 title = try_get(
2781 renderer, lambda x: x['title']['simpleText'], compat_str)
2782 yield self.url_result(
2783 'https://www.youtube.com/channel/%s' % channel_id,
2784 ie=YoutubeTabIE.ie_key(), video_title=title)
2785
2786 def _shelf_entries_from_content(self, shelf_renderer):
2787 content = shelf_renderer.get('content')
2788 if not isinstance(content, dict):
2789 return
2790 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
2791 if renderer:
2792 # TODO: add support for nested playlists so each shelf is processed
2793 # as separate playlist
2794 # TODO: this includes only first N items
2795 for entry in self._grid_entries(renderer):
2796 yield entry
2797 renderer = content.get('horizontalListRenderer')
2798 if renderer:
2799 # TODO
2800 pass
2801
2802 def _shelf_entries(self, shelf_renderer, skip_channels=False):
2803 ep = try_get(
2804 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
2805 compat_str)
2806 shelf_url = urljoin('https://www.youtube.com', ep)
2807 if shelf_url:
2808 # Skipping links to another channels, note that checking for
2809 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
2810 # will not work
2811 if skip_channels and '/channels?' in shelf_url:
2812 return
2813 title = try_get(
2814 shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2815 yield self.url_result(shelf_url, video_title=title)
2816 # Shelf may not contain shelf URL, fallback to extraction from content
2817 for entry in self._shelf_entries_from_content(shelf_renderer):
2818 yield entry
2819
2820 def _playlist_entries(self, video_list_renderer):
2821 for content in video_list_renderer['contents']:
2822 if not isinstance(content, dict):
2823 continue
2824 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
2825 if not isinstance(renderer, dict):
2826 continue
2827 video_id = renderer.get('videoId')
2828 if not video_id:
2829 continue
2830 yield self._extract_video(renderer)
2831
2832 def _rich_entries(self, rich_grid_renderer):
2833 renderer = try_get(
2834 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
2835 video_id = renderer.get('videoId')
2836 if not video_id:
2837 return
2838 yield self._extract_video(renderer)
2839
2840 def _video_entry(self, video_renderer):
2841 video_id = video_renderer.get('videoId')
2842 if video_id:
2843 return self._extract_video(video_renderer)
2844
2845 def _post_thread_entries(self, post_thread_renderer):
2846 post_renderer = try_get(
2847 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
2848 if not post_renderer:
2849 return
2850 # video attachment
2851 video_renderer = try_get(
2852 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict)
2853 video_id = None
2854 if video_renderer:
2855 entry = self._video_entry(video_renderer)
2856 if entry:
2857 yield entry
2858 # inline video links
2859 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
2860 for run in runs:
2861 if not isinstance(run, dict):
2862 continue
2863 ep_url = try_get(
2864 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
2865 if not ep_url:
2866 continue
2867 if not YoutubeIE.suitable(ep_url):
2868 continue
2869 ep_video_id = YoutubeIE._match_id(ep_url)
2870 if video_id == ep_video_id:
2871 continue
2872 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=video_id)
2873
2874 def _post_thread_continuation_entries(self, post_thread_continuation):
2875 contents = post_thread_continuation.get('contents')
2876 if not isinstance(contents, list):
2877 return
2878 for content in contents:
2879 renderer = content.get('backstagePostThreadRenderer')
2880 if not isinstance(renderer, dict):
2881 continue
2882 for entry in self._post_thread_entries(renderer):
2883 yield entry
2884
2885 r''' # unused
2886 def _rich_grid_entries(self, contents):
2887 for content in contents:
2888 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
2889 if video_renderer:
2890 entry = self._video_entry(video_renderer)
2891 if entry:
2892 yield entry
2893 '''
2894
2895 @staticmethod
2896 def _build_continuation_query(continuation, ctp=None):
2897 query = {
2898 'ctoken': continuation,
2899 'continuation': continuation,
2900 }
2901 if ctp:
2902 query['itct'] = ctp
2903 return query
2904
2905 @staticmethod
2906 def _extract_next_continuation_data(renderer):
2907 next_continuation = try_get(
2908 renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict)
2909 if not next_continuation:
2910 return
2911 continuation = next_continuation.get('continuation')
2912 if not continuation:
2913 return
2914 ctp = next_continuation.get('clickTrackingParams')
2915 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2916
2917 @classmethod
2918 def _extract_continuation(cls, renderer):
2919 next_continuation = cls._extract_next_continuation_data(renderer)
2920 if next_continuation:
2921 return next_continuation
2922 contents = []
2923 for key in ('contents', 'items'):
2924 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
2925 for content in contents:
2926 if not isinstance(content, dict):
2927 continue
2928 continuation_ep = try_get(
2929 content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
2930 dict)
2931 if not continuation_ep:
2932 continue
2933 continuation = try_get(
2934 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
2935 if not continuation:
2936 continue
2937 ctp = continuation_ep.get('clickTrackingParams')
2938 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2939
2940 def _entries(self, tab, item_id, identity_token, account_syncid):
2941
2942 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
2943 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
2944 for content in contents:
2945 if not isinstance(content, dict):
2946 continue
2947 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
2948 if not is_renderer:
2949 renderer = content.get('richItemRenderer')
2950 if renderer:
2951 for entry in self._rich_entries(renderer):
2952 yield entry
2953 continuation_list[0] = self._extract_continuation(parent_renderer)
2954 continue
2955 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
2956 for isr_content in isr_contents:
2957 if not isinstance(isr_content, dict):
2958 continue
2959
2960 known_renderers = {
2961 'playlistVideoListRenderer': self._playlist_entries,
2962 'gridRenderer': self._grid_entries,
2963 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'),
2964 'backstagePostThreadRenderer': self._post_thread_entries,
2965 'videoRenderer': lambda x: [self._video_entry(x)],
2966 }
2967 for key, renderer in isr_content.items():
2968 if key not in known_renderers:
2969 continue
2970 for entry in known_renderers[key](renderer):
2971 if entry:
2972 yield entry
2973 continuation_list[0] = self._extract_continuation(renderer)
2974 break
2975
2976 if not continuation_list[0]:
2977 continuation_list[0] = self._extract_continuation(is_renderer)
2978
2979 if not continuation_list[0]:
2980 continuation_list[0] = self._extract_continuation(parent_renderer)
2981
2982 continuation_list = [None] # Python 2 doesnot support nonlocal
2983 tab_content = try_get(tab, lambda x: x['content'], dict)
2984 if not tab_content:
2985 return
2986 parent_renderer = (
2987 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
2988 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
2989 for entry in extract_entries(parent_renderer):
2990 yield entry
2991 continuation = continuation_list[0]
2992
2993 headers = {
2994 'x-youtube-client-name': '1',
2995 'x-youtube-client-version': '2.20201112.04.01',
2996 }
2997 if identity_token:
2998 headers['x-youtube-identity-token'] = identity_token
2999
3000 if account_syncid:
3001 headers['X-Goog-PageId'] = account_syncid
3002 headers['X-Goog-AuthUser'] = 0
3003
3004 for page_num in itertools.count(1):
3005 if not continuation:
3006 break
3007 retries = self._downloader.params.get('extractor_retries', 3)
3008 count = -1
3009 last_error = None
3010 while count < retries:
3011 count += 1
3012 if last_error:
3013 self.report_warning('%s. Retrying ...' % last_error)
3014 try:
3015 response = self._call_api(
3016 ep='browse', fatal=True, headers=headers,
3017 video_id='%s page %s' % (item_id, page_num),
3018 query={
3019 'continuation': continuation['continuation'],
3020 'clickTracking': {'clickTrackingParams': continuation['itct']},
3021 },
3022 note='Downloading API JSON%s' % (' (retry #%d)' % count if count else ''))
3023 except ExtractorError as e:
3024 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404):
3025 # Downloading page may result in intermittent 5xx HTTP error
3026 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
3027 last_error = 'HTTP Error %s' % e.cause.code
3028 if count < retries:
3029 continue
3030 raise
3031 else:
3032 # Youtube sometimes sends incomplete data
3033 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
3034 if dict_get(response,
3035 ('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints')):
3036 break
3037
3038 # Youtube may send alerts if there was an issue with the continuation page
3039 self._extract_alerts(response, expected=False)
3040
3041 last_error = 'Incomplete data received'
3042 if count >= retries:
3043 self._downloader.report_error(last_error)
3044
3045 if not response:
3046 break
3047
3048 known_continuation_renderers = {
3049 'playlistVideoListContinuation': self._playlist_entries,
3050 'gridContinuation': self._grid_entries,
3051 'itemSectionContinuation': self._post_thread_continuation_entries,
3052 'sectionListContinuation': extract_entries, # for feeds
3053 }
3054 continuation_contents = try_get(
3055 response, lambda x: x['continuationContents'], dict) or {}
3056 continuation_renderer = None
3057 for key, value in continuation_contents.items():
3058 if key not in known_continuation_renderers:
3059 continue
3060 continuation_renderer = value
3061 continuation_list = [None]
3062 for entry in known_continuation_renderers[key](continuation_renderer):
3063 yield entry
3064 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
3065 break
3066 if continuation_renderer:
3067 continue
3068
3069 known_renderers = {
3070 'gridPlaylistRenderer': (self._grid_entries, 'items'),
3071 'gridVideoRenderer': (self._grid_entries, 'items'),
3072 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
3073 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
3074 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
3075 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
3076 }
3077 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3078 continuation_items = try_get(
3079 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
3080 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
3081 video_items_renderer = None
3082 for key, value in continuation_item.items():
3083 if key not in known_renderers:
3084 continue
3085 video_items_renderer = {known_renderers[key][1]: continuation_items}
3086 continuation_list = [None]
3087 for entry in known_renderers[key][0](video_items_renderer):
3088 yield entry
3089 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3090 break
3091 if video_items_renderer:
3092 continue
3093 break
3094
3095 @staticmethod
3096 def _extract_selected_tab(tabs):
3097 for tab in tabs:
3098 if try_get(tab, lambda x: x['tabRenderer']['selected'], bool):
3099 return tab['tabRenderer']
3100 else:
3101 raise ExtractorError('Unable to find selected tab')
3102
3103 @staticmethod
3104 def _extract_uploader(data):
3105 uploader = {}
3106 sidebar_renderer = try_get(
3107 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3108 if sidebar_renderer:
3109 for item in sidebar_renderer:
3110 if not isinstance(item, dict):
3111 continue
3112 renderer = item.get('playlistSidebarSecondaryInfoRenderer')
3113 if not isinstance(renderer, dict):
3114 continue
3115 owner = try_get(
3116 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3117 if owner:
3118 uploader['uploader'] = owner.get('text')
3119 uploader['uploader_id'] = try_get(
3120 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3121 uploader['uploader_url'] = urljoin(
3122 'https://www.youtube.com/',
3123 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3124 return {k: v for k, v in uploader.items() if v is not None}
3125
3126 def _extract_from_tabs(self, item_id, webpage, data, tabs):
3127 playlist_id = title = description = channel_url = channel_name = channel_id = None
3128 thumbnails_list = tags = []
3129
3130 selected_tab = self._extract_selected_tab(tabs)
3131 renderer = try_get(
3132 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3133 if renderer:
3134 channel_name = renderer.get('title')
3135 channel_url = renderer.get('channelUrl')
3136 channel_id = renderer.get('externalId')
3137 else:
3138 renderer = try_get(
3139 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3140
3141 if renderer:
3142 title = renderer.get('title')
3143 description = renderer.get('description', '')
3144 playlist_id = channel_id
3145 tags = renderer.get('keywords', '').split()
3146 thumbnails_list = (
3147 try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
3148 or try_get(
3149 data,
3150 lambda x: x['sidebar']['playlistSidebarRenderer']['items'][0]['playlistSidebarPrimaryInfoRenderer']['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
3151 list)
3152 or [])
3153
3154 thumbnails = []
3155 for t in thumbnails_list:
3156 if not isinstance(t, dict):
3157 continue
3158 thumbnail_url = url_or_none(t.get('url'))
3159 if not thumbnail_url:
3160 continue
3161 thumbnails.append({
3162 'url': thumbnail_url,
3163 'width': int_or_none(t.get('width')),
3164 'height': int_or_none(t.get('height')),
3165 })
3166 if playlist_id is None:
3167 playlist_id = item_id
3168 if title is None:
3169 title = (
3170 try_get(data, lambda x: x['header']['hashtagHeaderRenderer']['hashtag']['simpleText'])
3171 or playlist_id)
3172 title += format_field(selected_tab, 'title', ' - %s')
3173
3174 metadata = {
3175 'playlist_id': playlist_id,
3176 'playlist_title': title,
3177 'playlist_description': description,
3178 'uploader': channel_name,
3179 'uploader_id': channel_id,
3180 'uploader_url': channel_url,
3181 'thumbnails': thumbnails,
3182 'tags': tags,
3183 }
3184 if not channel_id:
3185 metadata.update(self._extract_uploader(data))
3186 metadata.update({
3187 'channel': metadata['uploader'],
3188 'channel_id': metadata['uploader_id'],
3189 'channel_url': metadata['uploader_url']})
3190 return self.playlist_result(
3191 self._entries(
3192 selected_tab, playlist_id,
3193 self._extract_identity_token(webpage, item_id),
3194 self._extract_account_syncid(data)),
3195 **metadata)
3196
3197 def _extract_mix_playlist(self, playlist, playlist_id):
3198 first_id = last_id = None
3199 for page_num in itertools.count(1):
3200 videos = list(self._playlist_entries(playlist))
3201 if not videos:
3202 return
3203 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
3204 if start >= len(videos):
3205 return
3206 for video in videos[start:]:
3207 if video['id'] == first_id:
3208 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
3209 return
3210 yield video
3211 first_id = first_id or videos[0]['id']
3212 last_id = videos[-1]['id']
3213
3214 _, data = self._extract_webpage(
3215 'https://www.youtube.com/watch?list=%s&v=%s' % (playlist_id, last_id),
3216 '%s page %d' % (playlist_id, page_num))
3217 playlist = try_get(
3218 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3219
3220 def _extract_from_playlist(self, item_id, url, data, playlist):
3221 title = playlist.get('title') or try_get(
3222 data, lambda x: x['titleText']['simpleText'], compat_str)
3223 playlist_id = playlist.get('playlistId') or item_id
3224
3225 # Delegating everything except mix playlists to regular tab-based playlist URL
3226 playlist_url = urljoin(url, try_get(
3227 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3228 compat_str))
3229 if playlist_url and playlist_url != url:
3230 return self.url_result(
3231 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3232 video_title=title)
3233
3234 return self.playlist_result(
3235 self._extract_mix_playlist(playlist, playlist_id),
3236 playlist_id=playlist_id, playlist_title=title)
3237
3238 def _extract_alerts(self, data, expected=False):
3239
3240 def _real_extract_alerts():
3241 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
3242 if not isinstance(alert_dict, dict):
3243 continue
3244 for alert in alert_dict.values():
3245 alert_type = alert.get('type')
3246 if not alert_type:
3247 continue
3248 message = try_get(alert, lambda x: x['text']['simpleText'], compat_str) or ''
3249 if message:
3250 yield alert_type, message
3251 for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
3252 message += try_get(run, lambda x: x['text'], compat_str)
3253 if message:
3254 yield alert_type, message
3255
3256 errors = []
3257 warnings = []
3258 for alert_type, alert_message in _real_extract_alerts():
3259 if alert_type.lower() == 'error':
3260 errors.append([alert_type, alert_message])
3261 else:
3262 warnings.append([alert_type, alert_message])
3263
3264 for alert_type, alert_message in (warnings + errors[:-1]):
3265 self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
3266 if errors:
3267 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
3268
3269 def _extract_webpage(self, url, item_id):
3270 retries = self._downloader.params.get('extractor_retries', 3)
3271 count = -1
3272 last_error = 'Incomplete yt initial data recieved'
3273 while count < retries:
3274 count += 1
3275 # Sometimes youtube returns a webpage with incomplete ytInitialData
3276 # See: https://github.com/yt-dlp/yt-dlp/issues/116
3277 if count:
3278 self.report_warning('%s. Retrying ...' % last_error)
3279 webpage = self._download_webpage(
3280 url, item_id,
3281 'Downloading webpage%s' % (' (retry #%d)' % count if count else ''))
3282 data = self._extract_yt_initial_data(item_id, webpage)
3283 self._extract_alerts(data, expected=True)
3284 if data.get('contents') or data.get('currentVideoEndpoint'):
3285 break
3286 if count >= retries:
3287 self._downloader.report_error(last_error)
3288 return webpage, data
3289
3290 def _real_extract(self, url):
3291 item_id = self._match_id(url)
3292 url = compat_urlparse.urlunparse(
3293 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
3294
3295 # This is not matched in a channel page with a tab selected
3296 mobj = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
3297 mobj = mobj.groupdict() if mobj else {}
3298 if mobj and not mobj.get('not_channel'):
3299 self._downloader.report_warning(
3300 'A channel/user page was given. All the channel\'s videos will be downloaded. '
3301 'To download only the videos in the home page, add a "/featured" to the URL')
3302 url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '')
3303
3304 # Handle both video/playlist URLs
3305 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3306 video_id = qs.get('v', [None])[0]
3307 playlist_id = qs.get('list', [None])[0]
3308
3309 if not video_id and (mobj.get('not_channel') or '').startswith('watch'):
3310 if not playlist_id:
3311 # If there is neither video or playlist ids,
3312 # youtube redirects to home page, which is undesirable
3313 raise ExtractorError('Unable to recognize tab page')
3314 self._downloader.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
3315 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
3316
3317 if video_id and playlist_id:
3318 if self._downloader.params.get('noplaylist'):
3319 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3320 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3321 self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
3322
3323 webpage, data = self._extract_webpage(url, item_id)
3324
3325 tabs = try_get(
3326 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
3327 if tabs:
3328 return self._extract_from_tabs(item_id, webpage, data, tabs)
3329
3330 playlist = try_get(
3331 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3332 if playlist:
3333 return self._extract_from_playlist(item_id, url, data, playlist)
3334
3335 video_id = try_get(
3336 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
3337 compat_str) or video_id
3338 if video_id:
3339 self._downloader.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
3340 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3341
3342 raise ExtractorError('Unable to recognize tab page')
3343
3344
3345 class YoutubePlaylistIE(InfoExtractor):
3346 IE_DESC = 'YouTube.com playlists'
3347 _VALID_URL = r'''(?x)(?:
3348 (?:https?://)?
3349 (?:\w+\.)?
3350 (?:
3351 (?:
3352 youtube(?:kids)?\.com|
3353 invidio\.us
3354 )
3355 /.*?\?.*?\blist=
3356 )?
3357 (?P<id>%(playlist_id)s)
3358 )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3359 IE_NAME = 'youtube:playlist'
3360 _TESTS = [{
3361 'note': 'issue #673',
3362 'url': 'PLBB231211A4F62143',
3363 'info_dict': {
3364 'title': '[OLD]Team Fortress 2 (Class-based LP)',
3365 'id': 'PLBB231211A4F62143',
3366 'uploader': 'Wickydoo',
3367 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
3368 },
3369 'playlist_mincount': 29,
3370 }, {
3371 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3372 'info_dict': {
3373 'title': 'YDL_safe_search',
3374 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3375 },
3376 'playlist_count': 2,
3377 'skip': 'This playlist is private',
3378 }, {
3379 'note': 'embedded',
3380 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3381 'playlist_count': 4,
3382 'info_dict': {
3383 'title': 'JODA15',
3384 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3385 'uploader': 'milan',
3386 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
3387 }
3388 }, {
3389 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3390 'playlist_mincount': 982,
3391 'info_dict': {
3392 'title': '2018 Chinese New Singles (11/6 updated)',
3393 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3394 'uploader': 'LBK',
3395 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
3396 }
3397 }, {
3398 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
3399 'only_matching': True,
3400 }, {
3401 # music album playlist
3402 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
3403 'only_matching': True,
3404 }]
3405
3406 @classmethod
3407 def suitable(cls, url):
3408 return False if YoutubeTabIE.suitable(url) else super(
3409 YoutubePlaylistIE, cls).suitable(url)
3410
3411 def _real_extract(self, url):
3412 playlist_id = self._match_id(url)
3413 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3414 if not qs:
3415 qs = {'list': playlist_id}
3416 return self.url_result(
3417 update_url_query('https://www.youtube.com/playlist', qs),
3418 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3419
3420
3421 class YoutubeYtBeIE(InfoExtractor):
3422 IE_DESC = 'youtu.be'
3423 _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}
3424 _TESTS = [{
3425 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
3426 'info_dict': {
3427 'id': 'yeWKywCrFtk',
3428 'ext': 'mp4',
3429 'title': 'Small Scale Baler and Braiding Rugs',
3430 'uploader': 'Backus-Page House Museum',
3431 'uploader_id': 'backuspagemuseum',
3432 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
3433 'upload_date': '20161008',
3434 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
3435 'categories': ['Nonprofits & Activism'],
3436 'tags': list,
3437 'like_count': int,
3438 'dislike_count': int,
3439 },
3440 'params': {
3441 'noplaylist': True,
3442 'skip_download': True,
3443 },
3444 }, {
3445 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
3446 'only_matching': True,
3447 }]
3448
3449 def _real_extract(self, url):
3450 mobj = re.match(self._VALID_URL, url)
3451 video_id = mobj.group('id')
3452 playlist_id = mobj.group('playlist_id')
3453 return self.url_result(
3454 update_url_query('https://www.youtube.com/watch', {
3455 'v': video_id,
3456 'list': playlist_id,
3457 'feature': 'youtu.be',
3458 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3459
3460
3461 class YoutubeYtUserIE(InfoExtractor):
3462 IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword'
3463 _VALID_URL = r'ytuser:(?P<id>.+)'
3464 _TESTS = [{
3465 'url': 'ytuser:phihag',
3466 'only_matching': True,
3467 }]
3468
3469 def _real_extract(self, url):
3470 user_id = self._match_id(url)
3471 return self.url_result(
3472 'https://www.youtube.com/user/%s' % user_id,
3473 ie=YoutubeTabIE.ie_key(), video_id=user_id)
3474
3475
3476 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3477 IE_NAME = 'youtube:favorites'
3478 IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)'
3479 _VALID_URL = r':ytfav(?:ou?rite)?s?'
3480 _LOGIN_REQUIRED = True
3481 _TESTS = [{
3482 'url': ':ytfav',
3483 'only_matching': True,
3484 }, {
3485 'url': ':ytfavorites',
3486 'only_matching': True,
3487 }]
3488
3489 def _real_extract(self, url):
3490 return self.url_result(
3491 'https://www.youtube.com/playlist?list=LL',
3492 ie=YoutubeTabIE.ie_key())
3493
3494
3495 class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
3496 IE_DESC = 'YouTube.com searches, "ytsearch" keyword'
3497 # there doesn't appear to be a real limit, for example if you search for
3498 # 'python' you get more than 8.000.000 results
3499 _MAX_RESULTS = float('inf')
3500 IE_NAME = 'youtube:search'
3501 _SEARCH_KEY = 'ytsearch'
3502 _SEARCH_PARAMS = None
3503 _TESTS = []
3504
3505 def _entries(self, query, n):
3506 data = {'query': query}
3507 if self._SEARCH_PARAMS:
3508 data['params'] = self._SEARCH_PARAMS
3509 total = 0
3510 for page_num in itertools.count(1):
3511 search = self._call_api(
3512 ep='search', video_id='query "%s"' % query, fatal=False,
3513 note='Downloading page %s' % page_num, query=data)
3514 if not search:
3515 break
3516 slr_contents = try_get(
3517 search,
3518 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
3519 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
3520 list)
3521 if not slr_contents:
3522 break
3523
3524 # Youtube sometimes adds promoted content to searches,
3525 # changing the index location of videos and token.
3526 # So we search through all entries till we find them.
3527 continuation_token = None
3528 for slr_content in slr_contents:
3529 if continuation_token is None:
3530 continuation_token = try_get(
3531 slr_content,
3532 lambda x: x['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'],
3533 compat_str)
3534
3535 isr_contents = try_get(
3536 slr_content,
3537 lambda x: x['itemSectionRenderer']['contents'],
3538 list)
3539 if not isr_contents:
3540 continue
3541 for content in isr_contents:
3542 if not isinstance(content, dict):
3543 continue
3544 video = content.get('videoRenderer')
3545 if not isinstance(video, dict):
3546 continue
3547 video_id = video.get('videoId')
3548 if not video_id:
3549 continue
3550
3551 yield self._extract_video(video)
3552 total += 1
3553 if total == n:
3554 return
3555
3556 if not continuation_token:
3557 break
3558 data['continuation'] = continuation_token
3559
3560 def _get_n_results(self, query, n):
3561 """Get a specified number of results for a query"""
3562 return self.playlist_result(self._entries(query, n), query)
3563
3564
3565 class YoutubeSearchDateIE(YoutubeSearchIE):
3566 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3567 _SEARCH_KEY = 'ytsearchdate'
3568 IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword'
3569 _SEARCH_PARAMS = 'CAI%3D'
3570
3571
3572 class YoutubeSearchURLIE(YoutubeSearchIE):
3573 IE_DESC = 'YouTube.com search URLs'
3574 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
3575 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
3576 # _MAX_RESULTS = 100
3577 _TESTS = [{
3578 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3579 'playlist_mincount': 5,
3580 'info_dict': {
3581 'title': 'youtube-dl test video',
3582 }
3583 }, {
3584 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3585 'only_matching': True,
3586 }]
3587
3588 @classmethod
3589 def _make_valid_url(cls):
3590 return cls._VALID_URL
3591
3592 def _real_extract(self, url):
3593 qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
3594 query = (qs.get('search_query') or qs.get('q'))[0]
3595 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
3596 return self._get_n_results(query, self._MAX_RESULTS)
3597
3598
3599 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
3600 """
3601 Base class for feed extractors
3602 Subclasses must define the _FEED_NAME property.
3603 """
3604 _LOGIN_REQUIRED = True
3605 _TESTS = []
3606
3607 @property
3608 def IE_NAME(self):
3609 return 'youtube:%s' % self._FEED_NAME
3610
3611 def _real_initialize(self):
3612 self._login()
3613
3614 def _real_extract(self, url):
3615 return self.url_result(
3616 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3617 ie=YoutubeTabIE.ie_key())
3618
3619
3620 class YoutubeWatchLaterIE(InfoExtractor):
3621 IE_NAME = 'youtube:watchlater'
3622 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3623 _VALID_URL = r':ytwatchlater'
3624 _TESTS = [{
3625 'url': ':ytwatchlater',
3626 'only_matching': True,
3627 }]
3628
3629 def _real_extract(self, url):
3630 return self.url_result(
3631 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
3632
3633
3634 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3635 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3636 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
3637 _FEED_NAME = 'recommended'
3638 _TESTS = [{
3639 'url': ':ytrec',
3640 'only_matching': True,
3641 }, {
3642 'url': ':ytrecommended',
3643 'only_matching': True,
3644 }, {
3645 'url': 'https://youtube.com',
3646 'only_matching': True,
3647 }]
3648
3649
3650 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3651 IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)'
3652 _VALID_URL = r':ytsub(?:scription)?s?'
3653 _FEED_NAME = 'subscriptions'
3654 _TESTS = [{
3655 'url': ':ytsubs',
3656 'only_matching': True,
3657 }, {
3658 'url': ':ytsubscriptions',
3659 'only_matching': True,
3660 }]
3661
3662
3663 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3664 IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)'
3665 _VALID_URL = r':ythis(?:tory)?'
3666 _FEED_NAME = 'history'
3667 _TESTS = [{
3668 'url': ':ythistory',
3669 'only_matching': True,
3670 }]
3671
3672
3673 class YoutubeTruncatedURLIE(InfoExtractor):
3674 IE_NAME = 'youtube:truncated_url'
3675 IE_DESC = False # Do not list
3676 _VALID_URL = r'''(?x)
3677 (?:https?://)?
3678 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3679 (?:watch\?(?:
3680 feature=[a-z_]+|
3681 annotation_id=annotation_[^&]+|
3682 x-yt-cl=[0-9]+|
3683 hl=[^&]*|
3684 t=[0-9]+
3685 )?
3686 |
3687 attribution_link\?a=[^&]+
3688 )
3689 $
3690 '''
3691
3692 _TESTS = [{
3693 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3694 'only_matching': True,
3695 }, {
3696 'url': 'https://www.youtube.com/watch?',
3697 'only_matching': True,
3698 }, {
3699 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3700 'only_matching': True,
3701 }, {
3702 'url': 'https://www.youtube.com/watch?feature=foo',
3703 'only_matching': True,
3704 }, {
3705 'url': 'https://www.youtube.com/watch?hl=en-GB',
3706 'only_matching': True,
3707 }, {
3708 'url': 'https://www.youtube.com/watch?t=2372',
3709 'only_matching': True,
3710 }]
3711
3712 def _real_extract(self, url):
3713 raise ExtractorError(
3714 'Did you forget to quote the URL? Remember that & is a meta '
3715 'character in most shells, so you want to put the URL in quotes, '
3716 'like youtube-dl '
3717 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3718 ' or simply youtube-dl BaW_jenozKc .',
3719 expected=True)
3720
3721
3722 class YoutubeTruncatedIDIE(InfoExtractor):
3723 IE_NAME = 'youtube:truncated_id'
3724 IE_DESC = False # Do not list
3725 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3726
3727 _TESTS = [{
3728 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3729 'only_matching': True,
3730 }]
3731
3732 def _real_extract(self, url):
3733 video_id = self._match_id(url)
3734 raise ExtractorError(
3735 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3736 expected=True)
fork of https://github.com/yt-dlp/yt-dlp/