]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/extractor/youtube.py
[YouTube] Show premium state in `availability` (#209)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
... / ...
CommitLineData
1# coding: utf-8
2
3from __future__ import unicode_literals
4
5import hashlib
6import itertools
7import json
8import os.path
9import random
10import re
11import time
12import traceback
13
14from .common import InfoExtractor, SearchInfoExtractor
15from ..compat import (
16 compat_chr,
17 compat_HTTPError,
18 compat_parse_qs,
19 compat_str,
20 compat_urllib_parse_unquote_plus,
21 compat_urllib_parse_urlencode,
22 compat_urllib_parse_urlparse,
23 compat_urlparse,
24)
25from ..jsinterp import JSInterpreter
26from ..utils import (
27 bool_or_none,
28 clean_html,
29 dict_get,
30 ExtractorError,
31 format_field,
32 float_or_none,
33 int_or_none,
34 mimetype2ext,
35 parse_codecs,
36 parse_duration,
37 qualities,
38 remove_start,
39 smuggle_url,
40 str_or_none,
41 str_to_int,
42 try_get,
43 unescapeHTML,
44 unified_strdate,
45 unsmuggle_url,
46 update_url_query,
47 url_or_none,
48 urlencode_postdata,
49 urljoin,
50)
51
52
53class YoutubeBaseInfoExtractor(InfoExtractor):
54 """Provide base functions for Youtube extractors"""
55 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
56 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
57
58 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
59 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
60 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
61
62 _RESERVED_NAMES = (
63 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|'
64 r'movies|results|shared|hashtag|trending|feed|feeds|'
65 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
66
67 _NETRC_MACHINE = 'youtube'
68 # If True it will raise an error if no login info is provided
69 _LOGIN_REQUIRED = False
70
71 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
72
73 def _ids_to_results(self, ids):
74 return [
75 self.url_result(vid_id, 'Youtube', video_id=vid_id)
76 for vid_id in ids]
77
78 def _login(self):
79 """
80 Attempt to log in to YouTube.
81 True is returned if successful or skipped.
82 False is returned if login failed.
83
84 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
85 """
86 username, password = self._get_login_info()
87 # No authentication to be performed
88 if username is None:
89 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
90 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
91 # if self._downloader.params.get('cookiefile'): # TODO remove 'and False' later - too many people using outdated cookies and open issues, remind them.
92 # self.to_screen('[Cookies] Reminder - Make sure to always use up to date cookies!')
93 return True
94
95 login_page = self._download_webpage(
96 self._LOGIN_URL, None,
97 note='Downloading login page',
98 errnote='unable to fetch login page', fatal=False)
99 if login_page is False:
100 return
101
102 login_form = self._hidden_inputs(login_page)
103
104 def req(url, f_req, note, errnote):
105 data = login_form.copy()
106 data.update({
107 'pstMsg': 1,
108 'checkConnection': 'youtube',
109 'checkedDomains': 'youtube',
110 'hl': 'en',
111 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
112 'f.req': json.dumps(f_req),
113 'flowName': 'GlifWebSignIn',
114 'flowEntry': 'ServiceLogin',
115 # TODO: reverse actual botguard identifier generation algo
116 'bgRequest': '["identifier",""]',
117 })
118 return self._download_json(
119 url, None, note=note, errnote=errnote,
120 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
121 fatal=False,
122 data=urlencode_postdata(data), headers={
123 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
124 'Google-Accounts-XSRF': 1,
125 })
126
127 def warn(message):
128 self._downloader.report_warning(message)
129
130 lookup_req = [
131 username,
132 None, [], None, 'US', None, None, 2, False, True,
133 [
134 None, None,
135 [2, 1, None, 1,
136 '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',
137 None, [], 4],
138 1, [None, None, []], None, None, None, True
139 ],
140 username,
141 ]
142
143 lookup_results = req(
144 self._LOOKUP_URL, lookup_req,
145 'Looking up account info', 'Unable to look up account info')
146
147 if lookup_results is False:
148 return False
149
150 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
151 if not user_hash:
152 warn('Unable to extract user hash')
153 return False
154
155 challenge_req = [
156 user_hash,
157 None, 1, None, [1, None, None, None, [password, None, True]],
158 [
159 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],
160 1, [None, None, []], None, None, None, True
161 ]]
162
163 challenge_results = req(
164 self._CHALLENGE_URL, challenge_req,
165 'Logging in', 'Unable to log in')
166
167 if challenge_results is False:
168 return
169
170 login_res = try_get(challenge_results, lambda x: x[0][5], list)
171 if login_res:
172 login_msg = try_get(login_res, lambda x: x[5], compat_str)
173 warn(
174 'Unable to login: %s' % 'Invalid password'
175 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
176 return False
177
178 res = try_get(challenge_results, lambda x: x[0][-1], list)
179 if not res:
180 warn('Unable to extract result entry')
181 return False
182
183 login_challenge = try_get(res, lambda x: x[0][0], list)
184 if login_challenge:
185 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
186 if challenge_str == 'TWO_STEP_VERIFICATION':
187 # SEND_SUCCESS - TFA code has been successfully sent to phone
188 # QUOTA_EXCEEDED - reached the limit of TFA codes
189 status = try_get(login_challenge, lambda x: x[5], compat_str)
190 if status == 'QUOTA_EXCEEDED':
191 warn('Exceeded the limit of TFA codes, try later')
192 return False
193
194 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
195 if not tl:
196 warn('Unable to extract TL')
197 return False
198
199 tfa_code = self._get_tfa_info('2-step verification code')
200
201 if not tfa_code:
202 warn(
203 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
204 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
205 return False
206
207 tfa_code = remove_start(tfa_code, 'G-')
208
209 tfa_req = [
210 user_hash, None, 2, None,
211 [
212 9, None, None, None, None, None, None, None,
213 [None, tfa_code, True, 2]
214 ]]
215
216 tfa_results = req(
217 self._TFA_URL.format(tl), tfa_req,
218 'Submitting TFA code', 'Unable to submit TFA code')
219
220 if tfa_results is False:
221 return False
222
223 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
224 if tfa_res:
225 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
226 warn(
227 'Unable to finish TFA: %s' % 'Invalid TFA code'
228 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
229 return False
230
231 check_cookie_url = try_get(
232 tfa_results, lambda x: x[0][-1][2], compat_str)
233 else:
234 CHALLENGES = {
235 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
236 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
237 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
238 }
239 challenge = CHALLENGES.get(
240 challenge_str,
241 '%s returned error %s.' % (self.IE_NAME, challenge_str))
242 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
243 return False
244 else:
245 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
246
247 if not check_cookie_url:
248 warn('Unable to extract CheckCookie URL')
249 return False
250
251 check_cookie_results = self._download_webpage(
252 check_cookie_url, None, 'Checking cookie', fatal=False)
253
254 if check_cookie_results is False:
255 return False
256
257 if 'https://myaccount.google.com/' not in check_cookie_results:
258 warn('Unable to log in')
259 return False
260
261 return True
262
263 def _initialize_consent(self):
264 cookies = self._get_cookies('https://www.youtube.com/')
265 if cookies.get('__Secure-3PSID'):
266 return
267 consent_id = None
268 consent = cookies.get('CONSENT')
269 if consent:
270 if 'YES' in consent.value:
271 return
272 consent_id = self._search_regex(
273 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
274 if not consent_id:
275 consent_id = random.randint(100, 999)
276 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
277
278 def _real_initialize(self):
279 self._initialize_consent()
280 if self._downloader is None:
281 return
282 if not self._login():
283 return
284
285 _YT_WEB_CLIENT_VERSION = '2.20210301.08.00'
286 _DEFAULT_API_DATA = {
287 'context': {
288 'client': {
289 'clientName': 'WEB',
290 'clientVersion': _YT_WEB_CLIENT_VERSION,
291 }
292 },
293 }
294
295 _DEFAULT_BASIC_API_HEADERS = {
296 'X-YouTube-Client-Name': '1',
297 'X-YouTube-Client-Version': _YT_WEB_CLIENT_VERSION
298 }
299
300 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
301 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
302 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
303
304 def _generate_sapisidhash_header(self):
305 sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
306 if sapisid_cookie is None:
307 return
308 time_now = round(time.time())
309 sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
310 return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
311
312 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
313 note='Downloading API JSON', errnote='Unable to download API page'):
314 data = self._DEFAULT_API_DATA.copy()
315 data.update(query)
316 headers = headers or {}
317 headers.update({'content-type': 'application/json'})
318 auth = self._generate_sapisidhash_header()
319 if auth is not None:
320 headers.update({'Authorization': auth, 'X-Origin': 'https://www.youtube.com'})
321 return self._download_json(
322 'https://www.youtube.com/youtubei/v1/%s' % ep,
323 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
324 data=json.dumps(data).encode('utf8'), headers=headers,
325 query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'})
326
327 def _extract_yt_initial_data(self, video_id, webpage):
328 return self._parse_json(
329 self._search_regex(
330 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
331 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
332 video_id)
333
334 def _extract_identity_token(self, webpage, item_id):
335 ytcfg = self._extract_ytcfg(item_id, webpage)
336 if ytcfg:
337 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
338 if token:
339 return token
340 return self._search_regex(
341 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
342 'identity token', default=None)
343
344 @staticmethod
345 def _extract_account_syncid(data):
346 """Extract syncId required to download private playlists of secondary channels"""
347 sync_ids = (
348 try_get(data, lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'], compat_str)
349 or '').split("||")
350 if len(sync_ids) >= 2 and sync_ids[1]:
351 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
352 # and just "user_syncid||" for primary channel. We only want the channel_syncid
353 return sync_ids[0]
354
355 def _extract_ytcfg(self, video_id, webpage):
356 return self._parse_json(
357 self._search_regex(
358 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
359 default='{}'), video_id, fatal=False)
360
361 def _extract_video(self, renderer):
362 video_id = renderer.get('videoId')
363 title = try_get(
364 renderer,
365 (lambda x: x['title']['runs'][0]['text'],
366 lambda x: x['title']['simpleText']), compat_str)
367 description = try_get(
368 renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'],
369 compat_str)
370 duration = parse_duration(try_get(
371 renderer, lambda x: x['lengthText']['simpleText'], compat_str))
372 view_count_text = try_get(
373 renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
374 view_count = str_to_int(self._search_regex(
375 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
376 'view count', default=None))
377 uploader = try_get(
378 renderer,
379 (lambda x: x['ownerText']['runs'][0]['text'],
380 lambda x: x['shortBylineText']['runs'][0]['text']), compat_str)
381 return {
382 '_type': 'url_transparent',
383 'ie_key': YoutubeIE.ie_key(),
384 'id': video_id,
385 'url': video_id,
386 'title': title,
387 'description': description,
388 'duration': duration,
389 'view_count': view_count,
390 'uploader': uploader,
391 }
392
393
394class YoutubeIE(YoutubeBaseInfoExtractor):
395 IE_DESC = 'YouTube.com'
396 _INVIDIOUS_SITES = (
397 # invidious-redirect websites
398 r'(?:www\.)?redirect\.invidious\.io',
399 r'(?:(?:www|dev)\.)?invidio\.us',
400 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
401 r'(?:www\.)?invidious\.pussthecat\.org',
402 r'(?:www\.)?invidious\.048596\.xyz',
403 r'(?:www\.)?invidious\.zee\.li',
404 r'(?:www\.)?vid\.puffyan\.us',
405 r'(?:(?:www|au)\.)?ytprivate\.com',
406 r'(?:www\.)?invidious\.namazso\.eu',
407 r'(?:www\.)?invidious\.ethibox\.fr',
408 r'(?:www\.)?inv\.skyn3t\.in',
409 r'(?:www\.)?invidious\.himiko\.cloud',
410 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
411 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
412 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
413 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
414 # youtube-dl invidious instances list
415 r'(?:(?:www|no)\.)?invidiou\.sh',
416 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
417 r'(?:www\.)?invidious\.kabi\.tk',
418 r'(?:www\.)?invidious\.13ad\.de',
419 r'(?:www\.)?invidious\.mastodon\.host',
420 r'(?:www\.)?invidious\.zapashcanon\.fr',
421 r'(?:www\.)?invidious\.kavin\.rocks',
422 r'(?:www\.)?invidious\.tube',
423 r'(?:www\.)?invidiou\.site',
424 r'(?:www\.)?invidious\.site',
425 r'(?:www\.)?invidious\.xyz',
426 r'(?:www\.)?invidious\.nixnet\.xyz',
427 r'(?:www\.)?invidious\.drycat\.fr',
428 r'(?:www\.)?tube\.poal\.co',
429 r'(?:www\.)?tube\.connect\.cafe',
430 r'(?:www\.)?vid\.wxzm\.sx',
431 r'(?:www\.)?vid\.mint\.lgbt',
432 r'(?:www\.)?yewtu\.be',
433 r'(?:www\.)?yt\.elukerio\.org',
434 r'(?:www\.)?yt\.lelux\.fi',
435 r'(?:www\.)?invidious\.ggc-project\.de',
436 r'(?:www\.)?yt\.maisputain\.ovh',
437 r'(?:www\.)?invidious\.toot\.koeln',
438 r'(?:www\.)?invidious\.fdn\.fr',
439 r'(?:www\.)?watch\.nettohikari\.com',
440 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
441 r'(?:www\.)?qklhadlycap4cnod\.onion',
442 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
443 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
444 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
445 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
446 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
447 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
448 )
449 _VALID_URL = r"""(?x)^
450 (
451 (?:https?://|//) # http(s):// or protocol-independent URL
452 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
453 (?:www\.)?deturl\.com/www\.youtube\.com|
454 (?:www\.)?pwnyoutube\.com|
455 (?:www\.)?hooktube\.com|
456 (?:www\.)?yourepeat\.com|
457 tube\.majestyc\.net|
458 %(invidious)s|
459 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
460 (?:.*?\#/)? # handle anchor (#/) redirect urls
461 (?: # the various things that can precede the ID:
462 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
463 |(?: # or the v= param in all its forms
464 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
465 (?:\?|\#!?) # the params delimiter ? or # or #!
466 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
467 v=
468 )
469 ))
470 |(?:
471 youtu\.be| # just youtu.be/xxxx
472 vid\.plus| # or vid.plus/xxxx
473 zwearz\.com/watch| # or zwearz.com/watch/xxxx
474 %(invidious)s
475 )/
476 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
477 )
478 )? # all until now is optional -> you can pass the naked ID
479 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
480 (?!.*?\blist=
481 (?:
482 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
483 WL # WL are handled by the watch later IE
484 )
485 )
486 (?(1).+)? # if we found the ID, everything can follow
487 $""" % {
488 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
489 'invidious': '|'.join(_INVIDIOUS_SITES),
490 }
491 _PLAYER_INFO_RE = (
492 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
493 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
494 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
495 )
496 _formats = {
497 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
498 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
499 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
500 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
501 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
502 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
503 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
504 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
505 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
506 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
507 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
508 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
509 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
510 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
511 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
512 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
513 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
514 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
515
516
517 # 3D videos
518 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
519 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
520 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
521 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
522 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
523 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
524 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
525
526 # Apple HTTP Live Streaming
527 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
528 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
529 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
530 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
531 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
532 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
533 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
534 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
535
536 # DASH mp4 video
537 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
538 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
539 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
540 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
541 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
542 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
543 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
544 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
545 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
546 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
547 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
548 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
549
550 # Dash mp4 audio
551 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
552 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
553 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
554 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
555 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
556 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
557 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
558
559 # Dash webm
560 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
561 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
562 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
563 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
564 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
565 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
566 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
567 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
568 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
569 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
570 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
571 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
572 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
573 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
574 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
575 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
576 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
577 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
578 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
579 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
580 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
581 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
582
583 # Dash webm audio
584 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
585 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
586
587 # Dash webm audio with opus inside
588 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
589 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
590 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
591
592 # RTMP (unnamed)
593 '_rtmp': {'protocol': 'rtmp'},
594
595 # av01 video only formats sometimes served with "unknown" codecs
596 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
597 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
598 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
599 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
600 }
601 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
602
603 _GEO_BYPASS = False
604
605 IE_NAME = 'youtube'
606 _TESTS = [
607 {
608 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
609 'info_dict': {
610 'id': 'BaW_jenozKc',
611 'ext': 'mp4',
612 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
613 'uploader': 'Philipp Hagemeister',
614 'uploader_id': 'phihag',
615 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
616 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
617 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
618 'upload_date': '20121002',
619 '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 .',
620 'categories': ['Science & Technology'],
621 'tags': ['youtube-dl'],
622 'duration': 10,
623 'view_count': int,
624 'like_count': int,
625 'dislike_count': int,
626 'start_time': 1,
627 'end_time': 9,
628 }
629 },
630 {
631 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
632 'note': 'Embed-only video (#1746)',
633 'info_dict': {
634 'id': 'yZIXLfi8CZQ',
635 'ext': 'mp4',
636 'upload_date': '20120608',
637 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
638 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
639 'uploader': 'SET India',
640 'uploader_id': 'setindia',
641 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
642 'age_limit': 18,
643 },
644 'skip': 'Private video',
645 },
646 {
647 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
648 'note': 'Use the first video ID in the URL',
649 'info_dict': {
650 'id': 'BaW_jenozKc',
651 'ext': 'mp4',
652 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
653 'uploader': 'Philipp Hagemeister',
654 'uploader_id': 'phihag',
655 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
656 'upload_date': '20121002',
657 '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 .',
658 'categories': ['Science & Technology'],
659 'tags': ['youtube-dl'],
660 'duration': 10,
661 'view_count': int,
662 'like_count': int,
663 'dislike_count': int,
664 },
665 'params': {
666 'skip_download': True,
667 },
668 },
669 {
670 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
671 'note': '256k DASH audio (format 141) via DASH manifest',
672 'info_dict': {
673 'id': 'a9LDPn-MO4I',
674 'ext': 'm4a',
675 'upload_date': '20121002',
676 'uploader_id': '8KVIDEO',
677 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
678 'description': '',
679 'uploader': '8KVIDEO',
680 'title': 'UHDTV TEST 8K VIDEO.mp4'
681 },
682 'params': {
683 'youtube_include_dash_manifest': True,
684 'format': '141',
685 },
686 'skip': 'format 141 not served anymore',
687 },
688 # DASH manifest with encrypted signature
689 {
690 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
691 'info_dict': {
692 'id': 'IB3lcPjvWLA',
693 'ext': 'm4a',
694 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
695 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
696 'duration': 244,
697 'uploader': 'AfrojackVEVO',
698 'uploader_id': 'AfrojackVEVO',
699 'upload_date': '20131011',
700 'abr': 129.495,
701 },
702 'params': {
703 'youtube_include_dash_manifest': True,
704 'format': '141/bestaudio[ext=m4a]',
705 },
706 },
707 # Controversy video
708 {
709 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
710 'info_dict': {
711 'id': 'T4XJQO3qol8',
712 'ext': 'mp4',
713 'duration': 219,
714 'upload_date': '20100909',
715 'uploader': 'Amazing Atheist',
716 'uploader_id': 'TheAmazingAtheist',
717 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
718 'title': 'Burning Everyone\'s Koran',
719 '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',
720 }
721 },
722 # Normal age-gate video (embed allowed)
723 {
724 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
725 'info_dict': {
726 'id': 'HtVdAasjOgU',
727 'ext': 'mp4',
728 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
729 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
730 'duration': 142,
731 'uploader': 'The Witcher',
732 'uploader_id': 'WitcherGame',
733 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
734 'upload_date': '20140605',
735 'age_limit': 18,
736 },
737 },
738 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
739 # YouTube Red ad is not captured for creator
740 {
741 'url': '__2ABJjxzNo',
742 'info_dict': {
743 'id': '__2ABJjxzNo',
744 'ext': 'mp4',
745 'duration': 266,
746 'upload_date': '20100430',
747 'uploader_id': 'deadmau5',
748 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
749 'creator': 'deadmau5',
750 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
751 'uploader': 'deadmau5',
752 'title': 'Deadmau5 - Some Chords (HD)',
753 'alt_title': 'Some Chords',
754 },
755 'expected_warnings': [
756 'DASH manifest missing',
757 ]
758 },
759 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
760 {
761 'url': 'lqQg6PlCWgI',
762 'info_dict': {
763 'id': 'lqQg6PlCWgI',
764 'ext': 'mp4',
765 'duration': 6085,
766 'upload_date': '20150827',
767 'uploader_id': 'olympic',
768 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
769 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
770 'uploader': 'Olympic',
771 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
772 },
773 'params': {
774 'skip_download': 'requires avconv',
775 }
776 },
777 # Non-square pixels
778 {
779 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
780 'info_dict': {
781 'id': '_b-2C3KPAM0',
782 'ext': 'mp4',
783 'stretched_ratio': 16 / 9.,
784 'duration': 85,
785 'upload_date': '20110310',
786 'uploader_id': 'AllenMeow',
787 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
788 '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',
789 'uploader': '孫ᄋᄅ',
790 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
791 },
792 },
793 # url_encoded_fmt_stream_map is empty string
794 {
795 'url': 'qEJwOuvDf7I',
796 'info_dict': {
797 'id': 'qEJwOuvDf7I',
798 'ext': 'webm',
799 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
800 'description': '',
801 'upload_date': '20150404',
802 'uploader_id': 'spbelect',
803 'uploader': 'Наблюдатели Петербурга',
804 },
805 'params': {
806 'skip_download': 'requires avconv',
807 },
808 'skip': 'This live event has ended.',
809 },
810 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
811 {
812 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
813 'info_dict': {
814 'id': 'FIl7x6_3R5Y',
815 'ext': 'webm',
816 'title': 'md5:7b81415841e02ecd4313668cde88737a',
817 'description': 'md5:116377fd2963b81ec4ce64b542173306',
818 'duration': 220,
819 'upload_date': '20150625',
820 'uploader_id': 'dorappi2000',
821 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
822 'uploader': 'dorappi2000',
823 'formats': 'mincount:31',
824 },
825 'skip': 'not actual anymore',
826 },
827 # DASH manifest with segment_list
828 {
829 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
830 'md5': '8ce563a1d667b599d21064e982ab9e31',
831 'info_dict': {
832 'id': 'CsmdDsKjzN8',
833 'ext': 'mp4',
834 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
835 'uploader': 'Airtek',
836 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
837 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
838 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
839 },
840 'params': {
841 'youtube_include_dash_manifest': True,
842 'format': '135', # bestvideo
843 },
844 'skip': 'This live event has ended.',
845 },
846 {
847 # Multifeed videos (multiple cameras), URL is for Main Camera
848 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
849 'info_dict': {
850 'id': 'jvGDaLqkpTg',
851 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
852 'description': 'md5:e03b909557865076822aa169218d6a5d',
853 },
854 'playlist': [{
855 'info_dict': {
856 'id': 'jvGDaLqkpTg',
857 'ext': 'mp4',
858 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
859 'description': 'md5:e03b909557865076822aa169218d6a5d',
860 'duration': 10643,
861 'upload_date': '20161111',
862 'uploader': 'Team PGP',
863 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
864 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
865 },
866 }, {
867 'info_dict': {
868 'id': '3AKt1R1aDnw',
869 'ext': 'mp4',
870 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
871 'description': 'md5:e03b909557865076822aa169218d6a5d',
872 'duration': 10991,
873 'upload_date': '20161111',
874 'uploader': 'Team PGP',
875 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
876 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
877 },
878 }, {
879 'info_dict': {
880 'id': 'RtAMM00gpVc',
881 'ext': 'mp4',
882 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
883 'description': 'md5:e03b909557865076822aa169218d6a5d',
884 'duration': 10995,
885 'upload_date': '20161111',
886 'uploader': 'Team PGP',
887 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
888 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
889 },
890 }, {
891 'info_dict': {
892 'id': '6N2fdlP3C5U',
893 'ext': 'mp4',
894 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
895 'description': 'md5:e03b909557865076822aa169218d6a5d',
896 'duration': 10990,
897 'upload_date': '20161111',
898 'uploader': 'Team PGP',
899 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
900 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
901 },
902 }],
903 'params': {
904 'skip_download': True,
905 },
906 },
907 {
908 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
909 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
910 'info_dict': {
911 'id': 'gVfLd0zydlo',
912 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
913 },
914 'playlist_count': 2,
915 'skip': 'Not multifeed anymore',
916 },
917 {
918 'url': 'https://vid.plus/FlRa-iH7PGw',
919 'only_matching': True,
920 },
921 {
922 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
923 'only_matching': True,
924 },
925 {
926 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
927 # Also tests cut-off URL expansion in video description (see
928 # https://github.com/ytdl-org/youtube-dl/issues/1892,
929 # https://github.com/ytdl-org/youtube-dl/issues/8164)
930 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
931 'info_dict': {
932 'id': 'lsguqyKfVQg',
933 'ext': 'mp4',
934 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
935 'alt_title': 'Dark Walk - Position Music',
936 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
937 'duration': 133,
938 'upload_date': '20151119',
939 'uploader_id': 'IronSoulElf',
940 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
941 'uploader': 'IronSoulElf',
942 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
943 'track': 'Dark Walk - Position Music',
944 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
945 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
946 },
947 'params': {
948 'skip_download': True,
949 },
950 },
951 {
952 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
953 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
954 'only_matching': True,
955 },
956 {
957 # Video with yt:stretch=17:0
958 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
959 'info_dict': {
960 'id': 'Q39EVAstoRM',
961 'ext': 'mp4',
962 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
963 'description': 'md5:ee18a25c350637c8faff806845bddee9',
964 'upload_date': '20151107',
965 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
966 'uploader': 'CH GAMER DROID',
967 },
968 'params': {
969 'skip_download': True,
970 },
971 'skip': 'This video does not exist.',
972 },
973 {
974 # Video licensed under Creative Commons
975 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
976 'info_dict': {
977 'id': 'M4gD1WSo5mA',
978 'ext': 'mp4',
979 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
980 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
981 'duration': 721,
982 'upload_date': '20150127',
983 'uploader_id': 'BerkmanCenter',
984 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
985 'uploader': 'The Berkman Klein Center for Internet & Society',
986 'license': 'Creative Commons Attribution license (reuse allowed)',
987 },
988 'params': {
989 'skip_download': True,
990 },
991 },
992 {
993 # Channel-like uploader_url
994 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
995 'info_dict': {
996 'id': 'eQcmzGIKrzg',
997 'ext': 'mp4',
998 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
999 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1000 'duration': 4060,
1001 'upload_date': '20151119',
1002 'uploader': 'Bernie Sanders',
1003 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1004 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1005 'license': 'Creative Commons Attribution license (reuse allowed)',
1006 },
1007 'params': {
1008 'skip_download': True,
1009 },
1010 },
1011 {
1012 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1013 'only_matching': True,
1014 },
1015 {
1016 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1017 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1018 'only_matching': True,
1019 },
1020 {
1021 # Rental video preview
1022 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1023 'info_dict': {
1024 'id': 'uGpuVWrhIzE',
1025 'ext': 'mp4',
1026 'title': 'Piku - Trailer',
1027 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1028 'upload_date': '20150811',
1029 'uploader': 'FlixMatrix',
1030 'uploader_id': 'FlixMatrixKaravan',
1031 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1032 'license': 'Standard YouTube License',
1033 },
1034 'params': {
1035 'skip_download': True,
1036 },
1037 'skip': 'This video is not available.',
1038 },
1039 {
1040 # YouTube Red video with episode data
1041 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1042 'info_dict': {
1043 'id': 'iqKdEhx-dD4',
1044 'ext': 'mp4',
1045 'title': 'Isolation - Mind Field (Ep 1)',
1046 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1047 'duration': 2085,
1048 'upload_date': '20170118',
1049 'uploader': 'Vsauce',
1050 'uploader_id': 'Vsauce',
1051 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1052 'series': 'Mind Field',
1053 'season_number': 1,
1054 'episode_number': 1,
1055 },
1056 'params': {
1057 'skip_download': True,
1058 },
1059 'expected_warnings': [
1060 'Skipping DASH manifest',
1061 ],
1062 },
1063 {
1064 # The following content has been identified by the YouTube community
1065 # as inappropriate or offensive to some audiences.
1066 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1067 'info_dict': {
1068 'id': '6SJNVb0GnPI',
1069 'ext': 'mp4',
1070 'title': 'Race Differences in Intelligence',
1071 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1072 'duration': 965,
1073 'upload_date': '20140124',
1074 'uploader': 'New Century Foundation',
1075 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1076 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1077 },
1078 'params': {
1079 'skip_download': True,
1080 },
1081 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1082 },
1083 {
1084 # itag 212
1085 'url': '1t24XAntNCY',
1086 'only_matching': True,
1087 },
1088 {
1089 # geo restricted to JP
1090 'url': 'sJL6WA-aGkQ',
1091 'only_matching': True,
1092 },
1093 {
1094 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1095 'only_matching': True,
1096 },
1097 {
1098 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1099 'only_matching': True,
1100 },
1101 {
1102 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1103 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1104 'only_matching': True,
1105 },
1106 {
1107 # DRM protected
1108 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1109 'only_matching': True,
1110 },
1111 {
1112 # Video with unsupported adaptive stream type formats
1113 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1114 'info_dict': {
1115 'id': 'Z4Vy8R84T1U',
1116 'ext': 'mp4',
1117 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1118 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1119 'duration': 433,
1120 'upload_date': '20130923',
1121 'uploader': 'Amelia Putri Harwita',
1122 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1123 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1124 'formats': 'maxcount:10',
1125 },
1126 'params': {
1127 'skip_download': True,
1128 'youtube_include_dash_manifest': False,
1129 },
1130 'skip': 'not actual anymore',
1131 },
1132 {
1133 # Youtube Music Auto-generated description
1134 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1135 'info_dict': {
1136 'id': 'MgNrAu2pzNs',
1137 'ext': 'mp4',
1138 'title': 'Voyeur Girl',
1139 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1140 'upload_date': '20190312',
1141 'uploader': 'Stephen - Topic',
1142 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1143 'artist': 'Stephen',
1144 'track': 'Voyeur Girl',
1145 'album': 'it\'s too much love to know my dear',
1146 'release_date': '20190313',
1147 'release_year': 2019,
1148 },
1149 'params': {
1150 'skip_download': True,
1151 },
1152 },
1153 {
1154 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1155 'only_matching': True,
1156 },
1157 {
1158 # invalid -> valid video id redirection
1159 'url': 'DJztXj2GPfl',
1160 'info_dict': {
1161 'id': 'DJztXj2GPfk',
1162 'ext': 'mp4',
1163 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1164 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1165 'upload_date': '20090125',
1166 'uploader': 'Prochorowka',
1167 'uploader_id': 'Prochorowka',
1168 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1169 'artist': 'Panjabi MC',
1170 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1171 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1172 },
1173 'params': {
1174 'skip_download': True,
1175 },
1176 'skip': 'Video unavailable',
1177 },
1178 {
1179 # empty description results in an empty string
1180 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1181 'info_dict': {
1182 'id': 'x41yOUIvK2k',
1183 'ext': 'mp4',
1184 'title': 'IMG 3456',
1185 'description': '',
1186 'upload_date': '20170613',
1187 'uploader_id': 'ElevageOrVert',
1188 'uploader': 'ElevageOrVert',
1189 },
1190 'params': {
1191 'skip_download': True,
1192 },
1193 },
1194 {
1195 # with '};' inside yt initial data (see [1])
1196 # see [2] for an example with '};' inside ytInitialPlayerResponse
1197 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1198 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1199 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1200 'info_dict': {
1201 'id': 'CHqg6qOn4no',
1202 'ext': 'mp4',
1203 'title': 'Part 77 Sort a list of simple types in c#',
1204 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1205 'upload_date': '20130831',
1206 'uploader_id': 'kudvenkat',
1207 'uploader': 'kudvenkat',
1208 },
1209 'params': {
1210 'skip_download': True,
1211 },
1212 },
1213 {
1214 # another example of '};' in ytInitialData
1215 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1216 'only_matching': True,
1217 },
1218 {
1219 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1220 'only_matching': True,
1221 },
1222 {
1223 # https://github.com/ytdl-org/youtube-dl/pull/28094
1224 'url': 'OtqTfy26tG0',
1225 'info_dict': {
1226 'id': 'OtqTfy26tG0',
1227 'ext': 'mp4',
1228 'title': 'Burn Out',
1229 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1230 'upload_date': '20141120',
1231 'uploader': 'The Cinematic Orchestra - Topic',
1232 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1233 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1234 'artist': 'The Cinematic Orchestra',
1235 'track': 'Burn Out',
1236 'album': 'Every Day',
1237 'release_data': None,
1238 'release_year': None,
1239 },
1240 'params': {
1241 'skip_download': True,
1242 },
1243 },
1244 {
1245 # controversial video, only works with bpctr when authenticated with cookies
1246 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1247 'only_matching': True,
1248 },
1249 ]
1250
1251 def __init__(self, *args, **kwargs):
1252 super(YoutubeIE, self).__init__(*args, **kwargs)
1253 self._code_cache = {}
1254 self._player_cache = {}
1255
1256 def _signature_cache_id(self, example_sig):
1257 """ Return a string representation of a signature """
1258 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1259
1260 @classmethod
1261 def _extract_player_info(cls, player_url):
1262 for player_re in cls._PLAYER_INFO_RE:
1263 id_m = re.search(player_re, player_url)
1264 if id_m:
1265 break
1266 else:
1267 raise ExtractorError('Cannot identify player %r' % player_url)
1268 return id_m.group('id')
1269
1270 def _extract_signature_function(self, video_id, player_url, example_sig):
1271 player_id = self._extract_player_info(player_url)
1272
1273 # Read from filesystem cache
1274 func_id = 'js_%s_%s' % (
1275 player_id, self._signature_cache_id(example_sig))
1276 assert os.path.basename(func_id) == func_id
1277
1278 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1279 if cache_spec is not None:
1280 return lambda s: ''.join(s[i] for i in cache_spec)
1281
1282 if player_id not in self._code_cache:
1283 self._code_cache[player_id] = self._download_webpage(
1284 player_url, video_id,
1285 note='Downloading player ' + player_id,
1286 errnote='Download of %s failed' % player_url)
1287 code = self._code_cache[player_id]
1288 res = self._parse_sig_js(code)
1289
1290 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1291 cache_res = res(test_string)
1292 cache_spec = [ord(c) for c in cache_res]
1293
1294 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1295 return res
1296
1297 def _print_sig_code(self, func, example_sig):
1298 def gen_sig_code(idxs):
1299 def _genslice(start, end, step):
1300 starts = '' if start == 0 else str(start)
1301 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1302 steps = '' if step == 1 else (':%d' % step)
1303 return 's[%s%s%s]' % (starts, ends, steps)
1304
1305 step = None
1306 # Quelch pyflakes warnings - start will be set when step is set
1307 start = '(Never used)'
1308 for i, prev in zip(idxs[1:], idxs[:-1]):
1309 if step is not None:
1310 if i - prev == step:
1311 continue
1312 yield _genslice(start, prev, step)
1313 step = None
1314 continue
1315 if i - prev in [-1, 1]:
1316 step = i - prev
1317 start = prev
1318 continue
1319 else:
1320 yield 's[%d]' % prev
1321 if step is None:
1322 yield 's[%d]' % i
1323 else:
1324 yield _genslice(start, i, step)
1325
1326 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1327 cache_res = func(test_string)
1328 cache_spec = [ord(c) for c in cache_res]
1329 expr_code = ' + '.join(gen_sig_code(cache_spec))
1330 signature_id_tuple = '(%s)' % (
1331 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1332 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1333 ' return %s\n') % (signature_id_tuple, expr_code)
1334 self.to_screen('Extracted signature function:\n' + code)
1335
1336 def _parse_sig_js(self, jscode):
1337 funcname = self._search_regex(
1338 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1339 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1340 r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
1341 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
1342 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}\(a,\d+\)',
1343 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\(\s*""\s*\)',
1344 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1345 # Obsolete patterns
1346 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1347 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1348 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1349 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1350 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1351 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1352 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1353 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1354 jscode, 'Initial JS player signature function name', group='sig')
1355
1356 jsi = JSInterpreter(jscode)
1357 initial_function = jsi.extract_function(funcname)
1358 return lambda s: initial_function([s])
1359
1360 def _decrypt_signature(self, s, video_id, player_url):
1361 """Turn the encrypted s field into a working signature"""
1362
1363 if player_url is None:
1364 raise ExtractorError('Cannot decrypt signature without player_url')
1365
1366 if player_url.startswith('//'):
1367 player_url = 'https:' + player_url
1368 elif not re.match(r'https?://', player_url):
1369 player_url = compat_urlparse.urljoin(
1370 'https://www.youtube.com', player_url)
1371 try:
1372 player_id = (player_url, self._signature_cache_id(s))
1373 if player_id not in self._player_cache:
1374 func = self._extract_signature_function(
1375 video_id, player_url, s
1376 )
1377 self._player_cache[player_id] = func
1378 func = self._player_cache[player_id]
1379 if self._downloader.params.get('youtube_print_sig_code'):
1380 self._print_sig_code(func, s)
1381 return func(s)
1382 except Exception as e:
1383 tb = traceback.format_exc()
1384 raise ExtractorError(
1385 'Signature extraction failed: ' + tb, cause=e)
1386
1387 def _mark_watched(self, video_id, player_response):
1388 playback_url = url_or_none(try_get(
1389 player_response,
1390 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']))
1391 if not playback_url:
1392 return
1393 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1394 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1395
1396 # cpn generation algorithm is reverse engineered from base.js.
1397 # In fact it works even with dummy cpn.
1398 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1399 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1400
1401 qs.update({
1402 'ver': ['2'],
1403 'cpn': [cpn],
1404 })
1405 playback_url = compat_urlparse.urlunparse(
1406 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1407
1408 self._download_webpage(
1409 playback_url, video_id, 'Marking watched',
1410 'Unable to mark watched', fatal=False)
1411
1412 @staticmethod
1413 def _extract_urls(webpage):
1414 # Embedded YouTube player
1415 entries = [
1416 unescapeHTML(mobj.group('url'))
1417 for mobj in re.finditer(r'''(?x)
1418 (?:
1419 <iframe[^>]+?src=|
1420 data-video-url=|
1421 <embed[^>]+?src=|
1422 embedSWF\(?:\s*|
1423 <object[^>]+data=|
1424 new\s+SWFObject\(
1425 )
1426 (["\'])
1427 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1428 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1429 \1''', webpage)]
1430
1431 # lazyYT YouTube embed
1432 entries.extend(list(map(
1433 unescapeHTML,
1434 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1435
1436 # Wordpress "YouTube Video Importer" plugin
1437 matches = re.findall(r'''(?x)<div[^>]+
1438 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1439 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1440 entries.extend(m[-1] for m in matches)
1441
1442 return entries
1443
1444 @staticmethod
1445 def _extract_url(webpage):
1446 urls = YoutubeIE._extract_urls(webpage)
1447 return urls[0] if urls else None
1448
1449 @classmethod
1450 def extract_id(cls, url):
1451 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1452 if mobj is None:
1453 raise ExtractorError('Invalid URL: %s' % url)
1454 video_id = mobj.group(2)
1455 return video_id
1456
1457 def _extract_chapters_from_json(self, data, video_id, duration):
1458 chapters_list = try_get(
1459 data,
1460 lambda x: x['playerOverlays']
1461 ['playerOverlayRenderer']
1462 ['decoratedPlayerBarRenderer']
1463 ['decoratedPlayerBarRenderer']
1464 ['playerBar']
1465 ['chapteredPlayerBarRenderer']
1466 ['chapters'],
1467 list)
1468 if not chapters_list:
1469 return
1470
1471 def chapter_time(chapter):
1472 return float_or_none(
1473 try_get(
1474 chapter,
1475 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1476 int),
1477 scale=1000)
1478 chapters = []
1479 for next_num, chapter in enumerate(chapters_list, start=1):
1480 start_time = chapter_time(chapter)
1481 if start_time is None:
1482 continue
1483 end_time = (chapter_time(chapters_list[next_num])
1484 if next_num < len(chapters_list) else duration)
1485 if end_time is None:
1486 continue
1487 title = try_get(
1488 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1489 compat_str)
1490 chapters.append({
1491 'start_time': start_time,
1492 'end_time': end_time,
1493 'title': title,
1494 })
1495 return chapters
1496
1497 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
1498 return self._parse_json(self._search_regex(
1499 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
1500 regex), webpage, name, default='{}'), video_id, fatal=False)
1501
1502 @staticmethod
1503 def _join_text_entries(runs):
1504 text = None
1505 for run in runs:
1506 if not isinstance(run, dict):
1507 continue
1508 sub_text = try_get(run, lambda x: x['text'], compat_str)
1509 if sub_text:
1510 if not text:
1511 text = sub_text
1512 continue
1513 text += sub_text
1514 return text
1515
1516 def _extract_comment(self, comment_renderer, parent=None):
1517 comment_id = comment_renderer.get('commentId')
1518 if not comment_id:
1519 return
1520 comment_text_runs = try_get(comment_renderer, lambda x: x['contentText']['runs']) or []
1521 text = self._join_text_entries(comment_text_runs) or ''
1522 comment_time_text = try_get(comment_renderer, lambda x: x['publishedTimeText']['runs']) or []
1523 time_text = self._join_text_entries(comment_time_text)
1524
1525 author = try_get(comment_renderer, lambda x: x['authorText']['simpleText'], compat_str)
1526 author_id = try_get(comment_renderer,
1527 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
1528 votes = str_to_int(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
1529 lambda x: x['likeCount']), compat_str)) or 0
1530 author_thumbnail = try_get(comment_renderer,
1531 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
1532
1533 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
1534 is_liked = try_get(comment_renderer, lambda x: x['isLiked'], bool)
1535
1536 return {
1537 'id': comment_id,
1538 'text': text,
1539 # TODO: This should be parsed to timestamp
1540 'time_text': time_text,
1541 'like_count': votes,
1542 'is_favorited': is_liked,
1543 'author': author,
1544 'author_id': author_id,
1545 'author_thumbnail': author_thumbnail,
1546 'author_is_uploader': author_is_uploader,
1547 'parent': parent or 'root'
1548 }
1549
1550 def _comment_entries(self, root_continuation_data, identity_token, account_syncid,
1551 session_token_list, parent=None, comment_counts=None):
1552
1553 def extract_thread(parent_renderer):
1554 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
1555 if not parent:
1556 comment_counts[2] = 0
1557 for content in contents:
1558 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
1559 comment_renderer = try_get(
1560 comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
1561 content, (lambda x: x['commentRenderer'], dict))
1562
1563 if not comment_renderer:
1564 continue
1565 comment = self._extract_comment(comment_renderer, parent)
1566 if not comment:
1567 continue
1568 comment_counts[0] += 1
1569 yield comment
1570 # Attempt to get the replies
1571 comment_replies_renderer = try_get(
1572 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
1573
1574 if comment_replies_renderer:
1575 comment_counts[2] += 1
1576 comment_entries_iter = self._comment_entries(
1577 comment_replies_renderer, identity_token, account_syncid,
1578 parent=comment.get('id'), session_token_list=session_token_list,
1579 comment_counts=comment_counts)
1580
1581 for reply_comment in comment_entries_iter:
1582 yield reply_comment
1583
1584 if not comment_counts:
1585 # comment so far, est. total comments, current comment thread #
1586 comment_counts = [0, 0, 0]
1587 headers = self._DEFAULT_BASIC_API_HEADERS.copy()
1588
1589 # TODO: Generalize the download code with TabIE
1590 if identity_token:
1591 headers['x-youtube-identity-token'] = identity_token
1592
1593 if account_syncid:
1594 headers['X-Goog-PageId'] = account_syncid
1595 headers['X-Goog-AuthUser'] = 0
1596
1597 continuation = YoutubeTabIE._extract_continuation(root_continuation_data) # TODO
1598 first_continuation = False
1599 if parent is None:
1600 first_continuation = True
1601
1602 for page_num in itertools.count(0):
1603 if not continuation:
1604 break
1605 retries = self._downloader.params.get('extractor_retries', 3)
1606 count = -1
1607 last_error = None
1608
1609 while count < retries:
1610 count += 1
1611 if last_error:
1612 self.report_warning('%s. Retrying ...' % last_error)
1613 try:
1614 query = {
1615 'ctoken': continuation['ctoken'],
1616 'pbj': 1,
1617 'type': 'next',
1618 }
1619 if parent:
1620 query['action_get_comment_replies'] = 1
1621 else:
1622 query['action_get_comments'] = 1
1623
1624 comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
1625 if page_num == 0:
1626 if first_continuation:
1627 note_prefix = "Downloading initial comment continuation page"
1628 else:
1629 note_prefix = " Downloading comment reply thread %d %s" % (comment_counts[2], comment_prog_str)
1630 else:
1631 note_prefix = "%sDownloading comment%s page %d %s" % (
1632 " " if parent else "",
1633 ' replies' if parent else '',
1634 page_num,
1635 comment_prog_str)
1636
1637 browse = self._download_json(
1638 'https://www.youtube.com/comment_service_ajax', None,
1639 '%s %s' % (note_prefix, '(retry #%d)' % count if count else ''),
1640 headers=headers, query=query,
1641 data=urlencode_postdata({
1642 'session_token': session_token_list[0]
1643 }))
1644 except ExtractorError as e:
1645 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404, 413):
1646 if e.cause.code == 413:
1647 self.report_warning("Assumed end of comments (received HTTP Error 413)")
1648 return
1649 # Downloading page may result in intermittent 5xx HTTP error
1650 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
1651 last_error = 'HTTP Error %s' % e.cause.code
1652 if e.cause.code == 404:
1653 last_error = last_error + " (this API is probably deprecated)"
1654 if count < retries:
1655 continue
1656 raise
1657 else:
1658 session_token = try_get(browse, lambda x: x['xsrf_token'], compat_str)
1659 if session_token:
1660 session_token_list[0] = session_token
1661
1662 response = try_get(browse,
1663 (lambda x: x['response'],
1664 lambda x: x[1]['response'])) or {}
1665
1666 if response.get('continuationContents'):
1667 break
1668
1669 # YouTube sometimes gives reload: now json if something went wrong (e.g. bad auth)
1670 if browse.get('reload'):
1671 raise ExtractorError("Invalid or missing params in continuation request", expected=False)
1672
1673 # TODO: not tested, merged from old extractor
1674 err_msg = browse.get('externalErrorMessage')
1675 if err_msg:
1676 raise ExtractorError('YouTube said: %s' % err_msg, expected=False)
1677
1678 # Youtube sometimes sends incomplete data
1679 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1680 last_error = 'Incomplete data received'
1681 if count >= retries:
1682 self._downloader.report_error(last_error)
1683
1684 if not response:
1685 break
1686
1687 known_continuation_renderers = {
1688 'itemSectionContinuation': extract_thread,
1689 'commentRepliesContinuation': extract_thread
1690 }
1691
1692 # extract next root continuation from the results
1693 continuation_contents = try_get(
1694 response, lambda x: x['continuationContents'], dict) or {}
1695
1696 for key, value in continuation_contents.items():
1697 if key not in known_continuation_renderers:
1698 continue
1699 continuation_renderer = value
1700
1701 if first_continuation:
1702 first_continuation = False
1703 expected_comment_count = try_get(
1704 continuation_renderer,
1705 (lambda x: x['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'],
1706 lambda x: x['header']['commentsHeaderRenderer']['commentsCount']['runs'][0]['text']),
1707 compat_str)
1708
1709 if expected_comment_count:
1710 comment_counts[1] = str_to_int(expected_comment_count)
1711 self.to_screen("Downloading ~%d comments" % str_to_int(expected_comment_count))
1712 yield comment_counts[1]
1713
1714 # TODO: cli arg.
1715 # 1/True for newest, 0/False for popular (default)
1716 comment_sort_index = int(True)
1717 sort_continuation_renderer = try_get(
1718 continuation_renderer,
1719 lambda x: x['header']['commentsHeaderRenderer']['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems']
1720 [comment_sort_index]['continuation']['reloadContinuationData'], dict)
1721 # If this fails, the initial continuation page
1722 # starts off with popular anyways.
1723 if sort_continuation_renderer:
1724 continuation = YoutubeTabIE._build_continuation_query(
1725 continuation=sort_continuation_renderer.get('continuation'),
1726 ctp=sort_continuation_renderer.get('clickTrackingParams'))
1727 self.to_screen("Sorting comments by %s" % ('popular' if comment_sort_index == 0 else 'newest'))
1728 break
1729
1730 for entry in known_continuation_renderers[key](continuation_renderer):
1731 yield entry
1732
1733 continuation = YoutubeTabIE._extract_continuation(continuation_renderer) # TODO
1734 break
1735
1736 def _extract_comments(self, ytcfg, video_id, contents, webpage, xsrf_token):
1737 """Entry for comment extraction"""
1738 comments = []
1739 known_entry_comment_renderers = (
1740 'itemSectionRenderer',
1741 )
1742 estimated_total = 0
1743 for entry in contents:
1744 for key, renderer in entry.items():
1745 if key not in known_entry_comment_renderers:
1746 continue
1747
1748 comment_iter = self._comment_entries(
1749 renderer,
1750 identity_token=self._extract_identity_token(webpage, item_id=video_id),
1751 account_syncid=self._extract_account_syncid(ytcfg),
1752 session_token_list=[xsrf_token])
1753
1754 for comment in comment_iter:
1755 if isinstance(comment, int):
1756 estimated_total = comment
1757 continue
1758 comments.append(comment)
1759 break
1760 self.to_screen("Downloaded %d/%d comments" % (len(comments), estimated_total))
1761 return {
1762 'comments': comments,
1763 'comment_count': len(comments),
1764 }
1765
1766 def _real_extract(self, url):
1767 url, smuggled_data = unsmuggle_url(url, {})
1768 video_id = self._match_id(url)
1769 base_url = self.http_scheme() + '//www.youtube.com/'
1770 webpage_url = base_url + 'watch?v=' + video_id
1771 webpage = self._download_webpage(
1772 webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False)
1773
1774 player_response = None
1775 if webpage:
1776 player_response = self._extract_yt_initial_variable(
1777 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
1778 video_id, 'initial player response')
1779 if not player_response:
1780 player_response = self._call_api(
1781 'player', {'videoId': video_id}, video_id)
1782
1783 playability_status = player_response.get('playabilityStatus') or {}
1784 if playability_status.get('reason') == 'Sign in to confirm your age':
1785 pr = self._parse_json(try_get(compat_parse_qs(
1786 self._download_webpage(
1787 base_url + 'get_video_info', video_id,
1788 'Refetching age-gated info webpage',
1789 'unable to download video info webpage', query={
1790 'video_id': video_id,
1791 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1792 }, fatal=False)),
1793 lambda x: x['player_response'][0],
1794 compat_str) or '{}', video_id)
1795 if pr:
1796 player_response = pr
1797
1798 trailer_video_id = try_get(
1799 playability_status,
1800 lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'],
1801 compat_str)
1802 if trailer_video_id:
1803 return self.url_result(
1804 trailer_video_id, self.ie_key(), trailer_video_id)
1805
1806 def get_text(x):
1807 if not x:
1808 return
1809 return x.get('simpleText') or ''.join([r['text'] for r in x['runs']])
1810
1811 search_meta = (
1812 lambda x: self._html_search_meta(x, webpage, default=None)) \
1813 if webpage else lambda x: None
1814
1815 video_details = player_response.get('videoDetails') or {}
1816 microformat = try_get(
1817 player_response,
1818 lambda x: x['microformat']['playerMicroformatRenderer'],
1819 dict) or {}
1820 video_title = video_details.get('title') \
1821 or get_text(microformat.get('title')) \
1822 or search_meta(['og:title', 'twitter:title', 'title'])
1823 video_description = video_details.get('shortDescription')
1824
1825 if not smuggled_data.get('force_singlefeed', False):
1826 if not self._downloader.params.get('noplaylist'):
1827 multifeed_metadata_list = try_get(
1828 player_response,
1829 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1830 compat_str)
1831 if multifeed_metadata_list:
1832 entries = []
1833 feed_ids = []
1834 for feed in multifeed_metadata_list.split(','):
1835 # Unquote should take place before split on comma (,) since textual
1836 # fields may contain comma as well (see
1837 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1838 feed_data = compat_parse_qs(
1839 compat_urllib_parse_unquote_plus(feed))
1840
1841 def feed_entry(name):
1842 return try_get(
1843 feed_data, lambda x: x[name][0], compat_str)
1844
1845 feed_id = feed_entry('id')
1846 if not feed_id:
1847 continue
1848 feed_title = feed_entry('title')
1849 title = video_title
1850 if feed_title:
1851 title += ' (%s)' % feed_title
1852 entries.append({
1853 '_type': 'url_transparent',
1854 'ie_key': 'Youtube',
1855 'url': smuggle_url(
1856 base_url + 'watch?v=' + feed_data['id'][0],
1857 {'force_singlefeed': True}),
1858 'title': title,
1859 })
1860 feed_ids.append(feed_id)
1861 self.to_screen(
1862 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1863 % (', '.join(feed_ids), video_id))
1864 return self.playlist_result(
1865 entries, video_id, video_title, video_description)
1866 else:
1867 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1868
1869 formats = []
1870 itags = []
1871 itag_qualities = {}
1872 player_url = None
1873 q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'])
1874 streaming_data = player_response.get('streamingData') or {}
1875 streaming_formats = streaming_data.get('formats') or []
1876 streaming_formats.extend(streaming_data.get('adaptiveFormats') or [])
1877 for fmt in streaming_formats:
1878 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
1879 continue
1880
1881 itag = str_or_none(fmt.get('itag'))
1882 quality = fmt.get('quality')
1883 if itag and quality:
1884 itag_qualities[itag] = quality
1885 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
1886 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
1887 # number of fragment that would subsequently requested with (`&sq=N`)
1888 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
1889 continue
1890
1891 fmt_url = fmt.get('url')
1892 if not fmt_url:
1893 sc = compat_parse_qs(fmt.get('signatureCipher'))
1894 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
1895 encrypted_sig = try_get(sc, lambda x: x['s'][0])
1896 if not (sc and fmt_url and encrypted_sig):
1897 continue
1898 if not player_url:
1899 if not webpage:
1900 continue
1901 player_url = self._search_regex(
1902 r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"',
1903 webpage, 'player URL', fatal=False)
1904 if not player_url:
1905 continue
1906 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
1907 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
1908 fmt_url += '&' + sp + '=' + signature
1909
1910 if itag:
1911 itags.append(itag)
1912 tbr = float_or_none(
1913 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
1914 dct = {
1915 'asr': int_or_none(fmt.get('audioSampleRate')),
1916 'filesize': int_or_none(fmt.get('contentLength')),
1917 'format_id': itag,
1918 'format_note': fmt.get('qualityLabel') or quality,
1919 'fps': int_or_none(fmt.get('fps')),
1920 'height': int_or_none(fmt.get('height')),
1921 'quality': q(quality),
1922 'tbr': tbr,
1923 'url': fmt_url,
1924 'width': fmt.get('width'),
1925 }
1926 mimetype = fmt.get('mimeType')
1927 if mimetype:
1928 mobj = re.match(
1929 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype)
1930 if mobj:
1931 dct['ext'] = mimetype2ext(mobj.group(1))
1932 dct.update(parse_codecs(mobj.group(2)))
1933 no_audio = dct.get('acodec') == 'none'
1934 no_video = dct.get('vcodec') == 'none'
1935 if no_audio:
1936 dct['vbr'] = tbr
1937 if no_video:
1938 dct['abr'] = tbr
1939 if no_audio or no_video:
1940 dct['downloader_options'] = {
1941 # Youtube throttles chunks >~10M
1942 'http_chunk_size': 10485760,
1943 }
1944 if dct.get('ext'):
1945 dct['container'] = dct['ext'] + '_dash'
1946 formats.append(dct)
1947
1948 hls_manifest_url = streaming_data.get('hlsManifestUrl')
1949 if hls_manifest_url:
1950 for f in self._extract_m3u8_formats(
1951 hls_manifest_url, video_id, 'mp4', fatal=False):
1952 itag = self._search_regex(
1953 r'/itag/(\d+)', f['url'], 'itag', default=None)
1954 if itag:
1955 f['format_id'] = itag
1956 formats.append(f)
1957
1958 if self._downloader.params.get('youtube_include_dash_manifest', True):
1959 dash_manifest_url = streaming_data.get('dashManifestUrl')
1960 if dash_manifest_url:
1961 for f in self._extract_mpd_formats(
1962 dash_manifest_url, video_id, fatal=False):
1963 itag = f['format_id']
1964 if itag in itags:
1965 continue
1966 if itag in itag_qualities:
1967 # Not actually usefull since the sorting is already done with "quality,res,fps,codec"
1968 # but kept to maintain feature parity (and code similarity) with youtube-dl
1969 # Remove if this causes any issues with sorting in future
1970 f['quality'] = q(itag_qualities[itag])
1971 filesize = int_or_none(self._search_regex(
1972 r'/clen/(\d+)', f.get('fragment_base_url')
1973 or f['url'], 'file size', default=None))
1974 if filesize:
1975 f['filesize'] = filesize
1976 formats.append(f)
1977
1978 if not formats:
1979 if not self._downloader.params.get('allow_unplayable_formats') and streaming_data.get('licenseInfos'):
1980 raise ExtractorError(
1981 'This video is DRM protected.', expected=True)
1982 pemr = try_get(
1983 playability_status,
1984 lambda x: x['errorScreen']['playerErrorMessageRenderer'],
1985 dict) or {}
1986 reason = get_text(pemr.get('reason')) or playability_status.get('reason')
1987 subreason = pemr.get('subreason')
1988 if subreason:
1989 subreason = clean_html(get_text(subreason))
1990 if subreason == 'The uploader has not made this video available in your country.':
1991 countries = microformat.get('availableCountries')
1992 if not countries:
1993 regions_allowed = search_meta('regionsAllowed')
1994 countries = regions_allowed.split(',') if regions_allowed else None
1995 self.raise_geo_restricted(
1996 subreason, countries)
1997 reason += '\n' + subreason
1998 if reason:
1999 raise ExtractorError(reason, expected=True)
2000
2001 self._sort_formats(formats)
2002
2003 keywords = video_details.get('keywords') or []
2004 if not keywords and webpage:
2005 keywords = [
2006 unescapeHTML(m.group('content'))
2007 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
2008 for keyword in keywords:
2009 if keyword.startswith('yt:stretch='):
2010 w, h = keyword.split('=')[1].split(':')
2011 w, h = int(w), int(h)
2012 if w > 0 and h > 0:
2013 ratio = w / h
2014 for f in formats:
2015 if f.get('vcodec') != 'none':
2016 f['stretched_ratio'] = ratio
2017
2018 thumbnails = []
2019 for container in (video_details, microformat):
2020 for thumbnail in (try_get(
2021 container,
2022 lambda x: x['thumbnail']['thumbnails'], list) or []):
2023 thumbnail_url = thumbnail.get('url')
2024 if not thumbnail_url:
2025 continue
2026 thumbnails.append({
2027 'height': int_or_none(thumbnail.get('height')),
2028 'url': thumbnail_url,
2029 'width': int_or_none(thumbnail.get('width')),
2030 })
2031 if thumbnails:
2032 break
2033 else:
2034 thumbnail = search_meta(['og:image', 'twitter:image'])
2035 if thumbnail:
2036 thumbnails = [{'url': thumbnail}]
2037
2038 category = microformat.get('category') or search_meta('genre')
2039 channel_id = video_details.get('channelId') \
2040 or microformat.get('externalChannelId') \
2041 or search_meta('channelId')
2042 duration = int_or_none(
2043 video_details.get('lengthSeconds')
2044 or microformat.get('lengthSeconds')) \
2045 or parse_duration(search_meta('duration'))
2046 is_live = video_details.get('isLive')
2047 owner_profile_url = microformat.get('ownerProfileUrl')
2048
2049 info = {
2050 'id': video_id,
2051 'title': self._live_title(video_title) if is_live else video_title,
2052 'formats': formats,
2053 'thumbnails': thumbnails,
2054 'description': video_description,
2055 'upload_date': unified_strdate(
2056 microformat.get('uploadDate')
2057 or search_meta('uploadDate')),
2058 'uploader': video_details['author'],
2059 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
2060 'uploader_url': owner_profile_url,
2061 'channel_id': channel_id,
2062 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None,
2063 'duration': duration,
2064 'view_count': int_or_none(
2065 video_details.get('viewCount')
2066 or microformat.get('viewCount')
2067 or search_meta('interactionCount')),
2068 'average_rating': float_or_none(video_details.get('averageRating')),
2069 'age_limit': 18 if (
2070 microformat.get('isFamilySafe') is False
2071 or search_meta('isFamilyFriendly') == 'false'
2072 or search_meta('og:restrictions:age') == '18+') else 0,
2073 'webpage_url': webpage_url,
2074 'categories': [category] if category else None,
2075 'tags': keywords,
2076 'is_live': is_live,
2077 'playable_in_embed': playability_status.get('playableInEmbed'),
2078 'was_live': video_details.get('isLiveContent'),
2079 }
2080
2081 pctr = try_get(
2082 player_response,
2083 lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
2084 subtitles = {}
2085 if pctr:
2086 def process_language(container, base_url, lang_code, query):
2087 lang_subs = []
2088 for fmt in self._SUBTITLE_FORMATS:
2089 query.update({
2090 'fmt': fmt,
2091 })
2092 lang_subs.append({
2093 'ext': fmt,
2094 'url': update_url_query(base_url, query),
2095 })
2096 container[lang_code] = lang_subs
2097
2098 for caption_track in (pctr.get('captionTracks') or []):
2099 base_url = caption_track.get('baseUrl')
2100 if not base_url:
2101 continue
2102 if caption_track.get('kind') != 'asr':
2103 lang_code = caption_track.get('languageCode')
2104 if not lang_code:
2105 continue
2106 process_language(
2107 subtitles, base_url, lang_code, {})
2108 continue
2109 automatic_captions = {}
2110 for translation_language in (pctr.get('translationLanguages') or []):
2111 translation_language_code = translation_language.get('languageCode')
2112 if not translation_language_code:
2113 continue
2114 process_language(
2115 automatic_captions, base_url, translation_language_code,
2116 {'tlang': translation_language_code})
2117 info['automatic_captions'] = automatic_captions
2118 info['subtitles'] = subtitles
2119
2120 parsed_url = compat_urllib_parse_urlparse(url)
2121 for component in [parsed_url.fragment, parsed_url.query]:
2122 query = compat_parse_qs(component)
2123 for k, v in query.items():
2124 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
2125 d_k += '_time'
2126 if d_k not in info and k in s_ks:
2127 info[d_k] = parse_duration(query[k][0])
2128
2129 # Youtube Music Auto-generated description
2130 if video_description:
2131 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)
2132 if mobj:
2133 release_year = mobj.group('release_year')
2134 release_date = mobj.group('release_date')
2135 if release_date:
2136 release_date = release_date.replace('-', '')
2137 if not release_year:
2138 release_year = release_date[:4]
2139 info.update({
2140 'album': mobj.group('album'.strip()),
2141 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
2142 'track': mobj.group('track').strip(),
2143 'release_date': release_date,
2144 'release_year': int_or_none(release_year),
2145 })
2146
2147 initial_data = None
2148 if webpage:
2149 initial_data = self._extract_yt_initial_variable(
2150 webpage, self._YT_INITIAL_DATA_RE, video_id,
2151 'yt initial data')
2152 if not initial_data:
2153 initial_data = self._call_api(
2154 'next', {'videoId': video_id}, video_id, fatal=False)
2155
2156 if not is_live:
2157 try:
2158 # This will error if there is no livechat
2159 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2160 info['subtitles']['live_chat'] = [{
2161 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
2162 'video_id': video_id,
2163 'ext': 'json',
2164 'protocol': 'youtube_live_chat_replay',
2165 }]
2166 except (KeyError, IndexError, TypeError):
2167 pass
2168
2169 if initial_data:
2170 chapters = self._extract_chapters_from_json(
2171 initial_data, video_id, duration)
2172 if not chapters:
2173 for engagment_pannel in (initial_data.get('engagementPanels') or []):
2174 contents = try_get(
2175 engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'],
2176 list)
2177 if not contents:
2178 continue
2179
2180 def chapter_time(mmlir):
2181 return parse_duration(
2182 get_text(mmlir.get('timeDescription')))
2183
2184 chapters = []
2185 for next_num, content in enumerate(contents, start=1):
2186 mmlir = content.get('macroMarkersListItemRenderer') or {}
2187 start_time = chapter_time(mmlir)
2188 end_time = chapter_time(try_get(
2189 contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \
2190 if next_num < len(contents) else duration
2191 if start_time is None or end_time is None:
2192 continue
2193 chapters.append({
2194 'start_time': start_time,
2195 'end_time': end_time,
2196 'title': get_text(mmlir.get('title')),
2197 })
2198 if chapters:
2199 break
2200 if chapters:
2201 info['chapters'] = chapters
2202
2203 contents = try_get(
2204 initial_data,
2205 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
2206 list) or []
2207 for content in contents:
2208 vpir = content.get('videoPrimaryInfoRenderer')
2209 if vpir:
2210 stl = vpir.get('superTitleLink')
2211 if stl:
2212 stl = get_text(stl)
2213 if try_get(
2214 vpir,
2215 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
2216 info['location'] = stl
2217 else:
2218 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
2219 if mobj:
2220 info.update({
2221 'series': mobj.group(1),
2222 'season_number': int(mobj.group(2)),
2223 'episode_number': int(mobj.group(3)),
2224 })
2225 for tlb in (try_get(
2226 vpir,
2227 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
2228 list) or []):
2229 tbr = tlb.get('toggleButtonRenderer') or {}
2230 for getter, regex in [(
2231 lambda x: x['defaultText']['accessibility']['accessibilityData'],
2232 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
2233 lambda x: x['accessibility'],
2234 lambda x: x['accessibilityData']['accessibilityData'],
2235 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
2236 label = (try_get(tbr, getter, dict) or {}).get('label')
2237 if label:
2238 mobj = re.match(regex, label)
2239 if mobj:
2240 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
2241 break
2242 sbr_tooltip = try_get(
2243 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
2244 if sbr_tooltip:
2245 like_count, dislike_count = sbr_tooltip.split(' / ')
2246 info.update({
2247 'like_count': str_to_int(like_count),
2248 'dislike_count': str_to_int(dislike_count),
2249 })
2250 vsir = content.get('videoSecondaryInfoRenderer')
2251 if vsir:
2252 info['channel'] = get_text(try_get(
2253 vsir,
2254 lambda x: x['owner']['videoOwnerRenderer']['title'],
2255 dict))
2256 rows = try_get(
2257 vsir,
2258 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
2259 list) or []
2260 multiple_songs = False
2261 for row in rows:
2262 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2263 multiple_songs = True
2264 break
2265 for row in rows:
2266 mrr = row.get('metadataRowRenderer') or {}
2267 mrr_title = mrr.get('title')
2268 if not mrr_title:
2269 continue
2270 mrr_title = get_text(mrr['title'])
2271 mrr_contents_text = get_text(mrr['contents'][0])
2272 if mrr_title == 'License':
2273 info['license'] = mrr_contents_text
2274 elif not multiple_songs:
2275 if mrr_title == 'Album':
2276 info['album'] = mrr_contents_text
2277 elif mrr_title == 'Artist':
2278 info['artist'] = mrr_contents_text
2279 elif mrr_title == 'Song':
2280 info['track'] = mrr_contents_text
2281
2282 fallbacks = {
2283 'channel': 'uploader',
2284 'channel_id': 'uploader_id',
2285 'channel_url': 'uploader_url',
2286 }
2287 for to, frm in fallbacks.items():
2288 if not info.get(to):
2289 info[to] = info.get(frm)
2290
2291 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
2292 v = info.get(s_k)
2293 if v:
2294 info[d_k] = v
2295
2296 is_private = bool_or_none(video_details.get('isPrivate'))
2297 is_unlisted = bool_or_none(microformat.get('isUnlisted'))
2298 is_membersonly = None
2299 is_premium = None
2300 if initial_data and is_private is not None:
2301 is_membersonly = False
2302 is_premium = False
2303 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list)
2304 for content in contents or []:
2305 badges = try_get(content, lambda x: x['videoPrimaryInfoRenderer']['badges'], list)
2306 for badge in badges or []:
2307 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label']) or ''
2308 if label.lower() == 'members only':
2309 is_membersonly = True
2310 break
2311 elif label.lower() == 'premium':
2312 is_premium = True
2313 break
2314 if is_membersonly or is_premium:
2315 break
2316
2317 # TODO: Add this for playlists
2318 info['availability'] = self._availability(
2319 is_private=is_private,
2320 needs_premium=is_premium,
2321 needs_subscription=is_membersonly,
2322 needs_auth=info['age_limit'] >= 18,
2323 is_unlisted=None if is_private is None else is_unlisted)
2324
2325 # get xsrf for annotations or comments
2326 get_annotations = self._downloader.params.get('writeannotations', False)
2327 get_comments = self._downloader.params.get('getcomments', False)
2328 if get_annotations or get_comments:
2329 xsrf_token = None
2330 ytcfg = self._extract_ytcfg(video_id, webpage)
2331 if ytcfg:
2332 xsrf_token = try_get(ytcfg, lambda x: x['XSRF_TOKEN'], compat_str)
2333 if not xsrf_token:
2334 xsrf_token = self._search_regex(
2335 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>(?:(?!\2).)+)\2',
2336 webpage, 'xsrf token', group='xsrf_token', fatal=False)
2337
2338 # annotations
2339 if get_annotations:
2340 invideo_url = try_get(
2341 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2342 if xsrf_token and invideo_url:
2343 xsrf_field_name = None
2344 if ytcfg:
2345 xsrf_field_name = try_get(ytcfg, lambda x: x['XSRF_FIELD_NAME'], compat_str)
2346 if not xsrf_field_name:
2347 xsrf_field_name = self._search_regex(
2348 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2349 webpage, 'xsrf field name',
2350 group='xsrf_field_name', default='session_token')
2351 info['annotations'] = self._download_webpage(
2352 self._proto_relative_url(invideo_url),
2353 video_id, note='Downloading annotations',
2354 errnote='Unable to download video annotations', fatal=False,
2355 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2356
2357 if get_comments:
2358 info['__post_extractor'] = lambda: self._extract_comments(ytcfg, video_id, contents, webpage, xsrf_token)
2359
2360 self.mark_watched(video_id, player_response)
2361
2362 return info
2363
2364
2365class YoutubeTabIE(YoutubeBaseInfoExtractor):
2366 IE_DESC = 'YouTube.com tab'
2367 _VALID_URL = r'''(?x)
2368 https?://
2369 (?:\w+\.)?
2370 (?:
2371 youtube(?:kids)?\.com|
2372 invidio\.us
2373 )/
2374 (?:
2375 (?:channel|c|user)/|
2376 (?P<not_channel>
2377 feed/|hashtag/|
2378 (?:playlist|watch)\?.*?\blist=
2379 )|
2380 (?!(?:%s)\b) # Direct URLs
2381 )
2382 (?P<id>[^/?\#&]+)
2383 ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
2384 IE_NAME = 'youtube:tab'
2385
2386 _TESTS = [{
2387 # playlists, multipage
2388 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
2389 'playlist_mincount': 94,
2390 'info_dict': {
2391 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2392 'title': 'Игорь Клейнер - Playlists',
2393 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2394 'uploader': 'Игорь Клейнер',
2395 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2396 },
2397 }, {
2398 # playlists, multipage, different order
2399 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2400 'playlist_mincount': 94,
2401 'info_dict': {
2402 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2403 'title': 'Игорь Клейнер - Playlists',
2404 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2405 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2406 'uploader': 'Игорь Клейнер',
2407 },
2408 }, {
2409 # playlists, singlepage
2410 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2411 'playlist_mincount': 4,
2412 'info_dict': {
2413 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2414 'title': 'ThirstForScience - Playlists',
2415 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
2416 'uploader': 'ThirstForScience',
2417 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2418 }
2419 }, {
2420 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
2421 'only_matching': True,
2422 }, {
2423 # basic, single video playlist
2424 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2425 'info_dict': {
2426 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2427 'uploader': 'Sergey M.',
2428 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2429 'title': 'youtube-dl public playlist',
2430 },
2431 'playlist_count': 1,
2432 }, {
2433 # empty playlist
2434 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2435 'info_dict': {
2436 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2437 'uploader': 'Sergey M.',
2438 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2439 'title': 'youtube-dl empty playlist',
2440 },
2441 'playlist_count': 0,
2442 }, {
2443 # Home tab
2444 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
2445 'info_dict': {
2446 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2447 'title': 'lex will - Home',
2448 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2449 'uploader': 'lex will',
2450 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2451 },
2452 'playlist_mincount': 2,
2453 }, {
2454 # Videos tab
2455 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
2456 'info_dict': {
2457 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2458 'title': 'lex will - Videos',
2459 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2460 'uploader': 'lex will',
2461 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2462 },
2463 'playlist_mincount': 975,
2464 }, {
2465 # Videos tab, sorted by popular
2466 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
2467 'info_dict': {
2468 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2469 'title': 'lex will - Videos',
2470 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2471 'uploader': 'lex will',
2472 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2473 },
2474 'playlist_mincount': 199,
2475 }, {
2476 # Playlists tab
2477 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
2478 'info_dict': {
2479 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2480 'title': 'lex will - Playlists',
2481 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2482 'uploader': 'lex will',
2483 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2484 },
2485 'playlist_mincount': 17,
2486 }, {
2487 # Community tab
2488 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
2489 'info_dict': {
2490 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2491 'title': 'lex will - Community',
2492 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2493 'uploader': 'lex will',
2494 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2495 },
2496 'playlist_mincount': 18,
2497 }, {
2498 # Channels tab
2499 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
2500 'info_dict': {
2501 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2502 'title': 'lex will - Channels',
2503 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2504 'uploader': 'lex will',
2505 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2506 },
2507 'playlist_mincount': 12,
2508 }, {
2509 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2510 'only_matching': True,
2511 }, {
2512 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2513 'only_matching': True,
2514 }, {
2515 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2516 'only_matching': True,
2517 }, {
2518 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2519 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2520 'info_dict': {
2521 'title': '29C3: Not my department',
2522 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2523 'uploader': 'Christiaan008',
2524 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
2525 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
2526 },
2527 'playlist_count': 96,
2528 }, {
2529 'note': 'Large playlist',
2530 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2531 'info_dict': {
2532 'title': 'Uploads from Cauchemar',
2533 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2534 'uploader': 'Cauchemar',
2535 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
2536 },
2537 'playlist_mincount': 1123,
2538 }, {
2539 # even larger playlist, 8832 videos
2540 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
2541 'only_matching': True,
2542 }, {
2543 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2544 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2545 'info_dict': {
2546 'title': 'Uploads from Interstellar Movie',
2547 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2548 'uploader': 'Interstellar Movie',
2549 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
2550 },
2551 'playlist_mincount': 21,
2552 }, {
2553 # https://github.com/ytdl-org/youtube-dl/issues/21844
2554 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2555 'info_dict': {
2556 'title': 'Data Analysis with Dr Mike Pound',
2557 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2558 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
2559 'uploader': 'Computerphile',
2560 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
2561 },
2562 'playlist_mincount': 11,
2563 }, {
2564 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2565 'only_matching': True,
2566 }, {
2567 # Playlist URL that does not actually serve a playlist
2568 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2569 'info_dict': {
2570 'id': 'FqZTN594JQw',
2571 'ext': 'webm',
2572 'title': "Smiley's People 01 detective, Adventure Series, Action",
2573 'uploader': 'STREEM',
2574 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2575 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2576 'upload_date': '20150526',
2577 'license': 'Standard YouTube License',
2578 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2579 'categories': ['People & Blogs'],
2580 'tags': list,
2581 'view_count': int,
2582 'like_count': int,
2583 'dislike_count': int,
2584 },
2585 'params': {
2586 'skip_download': True,
2587 },
2588 'skip': 'This video is not available.',
2589 'add_ie': [YoutubeIE.ie_key()],
2590 }, {
2591 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2592 'only_matching': True,
2593 }, {
2594 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
2595 'only_matching': True,
2596 }, {
2597 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
2598 'info_dict': {
2599 'id': '9Auq9mYxFEE',
2600 'ext': 'mp4',
2601 'title': compat_str,
2602 'uploader': 'Sky News',
2603 'uploader_id': 'skynews',
2604 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
2605 'upload_date': '20191102',
2606 'description': 'md5:85ddd75d888674631aaf9599a9a0b0ae',
2607 'categories': ['News & Politics'],
2608 'tags': list,
2609 'like_count': int,
2610 'dislike_count': int,
2611 },
2612 'params': {
2613 'skip_download': True,
2614 },
2615 }, {
2616 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2617 'info_dict': {
2618 'id': 'a48o2S1cPoo',
2619 'ext': 'mp4',
2620 'title': 'The Young Turks - Live Main Show',
2621 'uploader': 'The Young Turks',
2622 'uploader_id': 'TheYoungTurks',
2623 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2624 'upload_date': '20150715',
2625 'license': 'Standard YouTube License',
2626 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2627 'categories': ['News & Politics'],
2628 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2629 'like_count': int,
2630 'dislike_count': int,
2631 },
2632 'params': {
2633 'skip_download': True,
2634 },
2635 'only_matching': True,
2636 }, {
2637 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2638 'only_matching': True,
2639 }, {
2640 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2641 'only_matching': True,
2642 }, {
2643 'url': 'https://www.youtube.com/feed/trending',
2644 'only_matching': True,
2645 }, {
2646 # needs auth
2647 'url': 'https://www.youtube.com/feed/library',
2648 'only_matching': True,
2649 }, {
2650 # needs auth
2651 'url': 'https://www.youtube.com/feed/history',
2652 'only_matching': True,
2653 }, {
2654 # needs auth
2655 'url': 'https://www.youtube.com/feed/subscriptions',
2656 'only_matching': True,
2657 }, {
2658 # needs auth
2659 'url': 'https://www.youtube.com/feed/watch_later',
2660 'only_matching': True,
2661 }, {
2662 # no longer available?
2663 'url': 'https://www.youtube.com/feed/recommended',
2664 'only_matching': True,
2665 }, {
2666 # inline playlist with not always working continuations
2667 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
2668 'only_matching': True,
2669 }, {
2670 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8',
2671 'only_matching': True,
2672 }, {
2673 'url': 'https://www.youtube.com/course',
2674 'only_matching': True,
2675 }, {
2676 'url': 'https://www.youtube.com/zsecurity',
2677 'only_matching': True,
2678 }, {
2679 'url': 'http://www.youtube.com/NASAgovVideo/videos',
2680 'only_matching': True,
2681 }, {
2682 'url': 'https://www.youtube.com/TheYoungTurks/live',
2683 'only_matching': True,
2684 }]
2685
2686 @classmethod
2687 def suitable(cls, url):
2688 return False if YoutubeIE.suitable(url) else super(
2689 YoutubeTabIE, cls).suitable(url)
2690
2691 def _extract_channel_id(self, webpage):
2692 channel_id = self._html_search_meta(
2693 'channelId', webpage, 'channel id', default=None)
2694 if channel_id:
2695 return channel_id
2696 channel_url = self._html_search_meta(
2697 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
2698 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
2699 'twitter:app:url:googleplay'), webpage, 'channel url')
2700 return self._search_regex(
2701 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
2702 channel_url, 'channel id')
2703
2704 @staticmethod
2705 def _extract_basic_item_renderer(item):
2706 # Modified from _extract_grid_item_renderer
2707 known_renderers = (
2708 'playlistRenderer', 'videoRenderer', 'channelRenderer',
2709 'gridPlaylistRenderer', 'gridVideoRenderer', 'gridChannelRenderer'
2710 )
2711 for key, renderer in item.items():
2712 if key not in known_renderers:
2713 continue
2714 return renderer
2715
2716 def _grid_entries(self, grid_renderer):
2717 for item in grid_renderer['items']:
2718 if not isinstance(item, dict):
2719 continue
2720 renderer = self._extract_basic_item_renderer(item)
2721 if not isinstance(renderer, dict):
2722 continue
2723 title = try_get(
2724 renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2725 # playlist
2726 playlist_id = renderer.get('playlistId')
2727 if playlist_id:
2728 yield self.url_result(
2729 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2730 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
2731 video_title=title)
2732 # video
2733 video_id = renderer.get('videoId')
2734 if video_id:
2735 yield self._extract_video(renderer)
2736 # channel
2737 channel_id = renderer.get('channelId')
2738 if channel_id:
2739 title = try_get(
2740 renderer, lambda x: x['title']['simpleText'], compat_str)
2741 yield self.url_result(
2742 'https://www.youtube.com/channel/%s' % channel_id,
2743 ie=YoutubeTabIE.ie_key(), video_title=title)
2744
2745 def _shelf_entries_from_content(self, shelf_renderer):
2746 content = shelf_renderer.get('content')
2747 if not isinstance(content, dict):
2748 return
2749 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
2750 if renderer:
2751 # TODO: add support for nested playlists so each shelf is processed
2752 # as separate playlist
2753 # TODO: this includes only first N items
2754 for entry in self._grid_entries(renderer):
2755 yield entry
2756 renderer = content.get('horizontalListRenderer')
2757 if renderer:
2758 # TODO
2759 pass
2760
2761 def _shelf_entries(self, shelf_renderer, skip_channels=False):
2762 ep = try_get(
2763 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
2764 compat_str)
2765 shelf_url = urljoin('https://www.youtube.com', ep)
2766 if shelf_url:
2767 # Skipping links to another channels, note that checking for
2768 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
2769 # will not work
2770 if skip_channels and '/channels?' in shelf_url:
2771 return
2772 title = try_get(
2773 shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2774 yield self.url_result(shelf_url, video_title=title)
2775 # Shelf may not contain shelf URL, fallback to extraction from content
2776 for entry in self._shelf_entries_from_content(shelf_renderer):
2777 yield entry
2778
2779 def _playlist_entries(self, video_list_renderer):
2780 for content in video_list_renderer['contents']:
2781 if not isinstance(content, dict):
2782 continue
2783 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
2784 if not isinstance(renderer, dict):
2785 continue
2786 video_id = renderer.get('videoId')
2787 if not video_id:
2788 continue
2789 yield self._extract_video(renderer)
2790
2791 def _rich_entries(self, rich_grid_renderer):
2792 renderer = try_get(
2793 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
2794 video_id = renderer.get('videoId')
2795 if not video_id:
2796 return
2797 yield self._extract_video(renderer)
2798
2799 def _video_entry(self, video_renderer):
2800 video_id = video_renderer.get('videoId')
2801 if video_id:
2802 return self._extract_video(video_renderer)
2803
2804 def _post_thread_entries(self, post_thread_renderer):
2805 post_renderer = try_get(
2806 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
2807 if not post_renderer:
2808 return
2809 # video attachment
2810 video_renderer = try_get(
2811 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict)
2812 video_id = None
2813 if video_renderer:
2814 entry = self._video_entry(video_renderer)
2815 if entry:
2816 yield entry
2817 # inline video links
2818 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
2819 for run in runs:
2820 if not isinstance(run, dict):
2821 continue
2822 ep_url = try_get(
2823 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
2824 if not ep_url:
2825 continue
2826 if not YoutubeIE.suitable(ep_url):
2827 continue
2828 ep_video_id = YoutubeIE._match_id(ep_url)
2829 if video_id == ep_video_id:
2830 continue
2831 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=video_id)
2832
2833 def _post_thread_continuation_entries(self, post_thread_continuation):
2834 contents = post_thread_continuation.get('contents')
2835 if not isinstance(contents, list):
2836 return
2837 for content in contents:
2838 renderer = content.get('backstagePostThreadRenderer')
2839 if not isinstance(renderer, dict):
2840 continue
2841 for entry in self._post_thread_entries(renderer):
2842 yield entry
2843
2844 @staticmethod
2845 def _build_continuation_query(continuation, ctp=None):
2846 query = {
2847 'ctoken': continuation,
2848 'continuation': continuation,
2849 }
2850 if ctp:
2851 query['itct'] = ctp
2852 return query
2853
2854 @staticmethod
2855 def _extract_next_continuation_data(renderer):
2856 next_continuation = try_get(
2857 renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict)
2858 if not next_continuation:
2859 return
2860 continuation = next_continuation.get('continuation')
2861 if not continuation:
2862 return
2863 ctp = next_continuation.get('clickTrackingParams')
2864 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2865
2866 @classmethod
2867 def _extract_continuation(cls, renderer):
2868 next_continuation = cls._extract_next_continuation_data(renderer)
2869 if next_continuation:
2870 return next_continuation
2871 contents = []
2872 for key in ('contents', 'items'):
2873 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
2874 for content in contents:
2875 if not isinstance(content, dict):
2876 continue
2877 continuation_ep = try_get(
2878 content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
2879 dict)
2880 if not continuation_ep:
2881 continue
2882 continuation = try_get(
2883 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
2884 if not continuation:
2885 continue
2886 ctp = continuation_ep.get('clickTrackingParams')
2887 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2888
2889 def _entries(self, tab, item_id, identity_token, account_syncid):
2890
2891 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
2892 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
2893 for content in contents:
2894 if not isinstance(content, dict):
2895 continue
2896 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
2897 if not is_renderer:
2898 renderer = content.get('richItemRenderer')
2899 if renderer:
2900 for entry in self._rich_entries(renderer):
2901 yield entry
2902 continuation_list[0] = self._extract_continuation(parent_renderer)
2903 continue
2904 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
2905 for isr_content in isr_contents:
2906 if not isinstance(isr_content, dict):
2907 continue
2908
2909 known_renderers = {
2910 'playlistVideoListRenderer': self._playlist_entries,
2911 'gridRenderer': self._grid_entries,
2912 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'),
2913 'backstagePostThreadRenderer': self._post_thread_entries,
2914 'videoRenderer': lambda x: [self._video_entry(x)],
2915 }
2916 for key, renderer in isr_content.items():
2917 if key not in known_renderers:
2918 continue
2919 for entry in known_renderers[key](renderer):
2920 if entry:
2921 yield entry
2922 continuation_list[0] = self._extract_continuation(renderer)
2923 break
2924
2925 if not continuation_list[0]:
2926 continuation_list[0] = self._extract_continuation(is_renderer)
2927
2928 if not continuation_list[0]:
2929 continuation_list[0] = self._extract_continuation(parent_renderer)
2930
2931 continuation_list = [None] # Python 2 doesnot support nonlocal
2932 tab_content = try_get(tab, lambda x: x['content'], dict)
2933 if not tab_content:
2934 return
2935 parent_renderer = (
2936 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
2937 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
2938 for entry in extract_entries(parent_renderer):
2939 yield entry
2940 continuation = continuation_list[0]
2941
2942 headers = {
2943 'x-youtube-client-name': '1',
2944 'x-youtube-client-version': '2.20201112.04.01',
2945 }
2946 if identity_token:
2947 headers['x-youtube-identity-token'] = identity_token
2948
2949 if account_syncid:
2950 headers['X-Goog-PageId'] = account_syncid
2951 headers['X-Goog-AuthUser'] = 0
2952
2953 for page_num in itertools.count(1):
2954 if not continuation:
2955 break
2956 retries = self._downloader.params.get('extractor_retries', 3)
2957 count = -1
2958 last_error = None
2959 while count < retries:
2960 count += 1
2961 if last_error:
2962 self.report_warning('%s. Retrying ...' % last_error)
2963 try:
2964 response = self._call_api(
2965 ep="browse", fatal=True, headers=headers,
2966 video_id='%s page %s' % (item_id, page_num),
2967 query={
2968 'continuation': continuation['continuation'],
2969 'clickTracking': {'clickTrackingParams': continuation['itct']},
2970 },
2971 note='Downloading API JSON%s' % (' (retry #%d)' % count if count else ''))
2972 except ExtractorError as e:
2973 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404):
2974 # Downloading page may result in intermittent 5xx HTTP error
2975 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
2976 last_error = 'HTTP Error %s' % e.cause.code
2977 if count < retries:
2978 continue
2979 raise
2980 else:
2981 # Youtube sometimes sends incomplete data
2982 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
2983 if dict_get(response,
2984 ('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints')):
2985 break
2986
2987 # Youtube may send alerts if there was an issue with the continuation page
2988 self._extract_alerts(response, expected=False)
2989
2990 last_error = 'Incomplete data received'
2991 if count >= retries:
2992 self._downloader.report_error(last_error)
2993
2994 if not response:
2995 break
2996
2997 known_continuation_renderers = {
2998 'playlistVideoListContinuation': self._playlist_entries,
2999 'gridContinuation': self._grid_entries,
3000 'itemSectionContinuation': self._post_thread_continuation_entries,
3001 'sectionListContinuation': extract_entries, # for feeds
3002 }
3003 continuation_contents = try_get(
3004 response, lambda x: x['continuationContents'], dict) or {}
3005 continuation_renderer = None
3006 for key, value in continuation_contents.items():
3007 if key not in known_continuation_renderers:
3008 continue
3009 continuation_renderer = value
3010 continuation_list = [None]
3011 for entry in known_continuation_renderers[key](continuation_renderer):
3012 yield entry
3013 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
3014 break
3015 if continuation_renderer:
3016 continue
3017
3018 known_renderers = {
3019 'gridPlaylistRenderer': (self._grid_entries, 'items'),
3020 'gridVideoRenderer': (self._grid_entries, 'items'),
3021 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
3022 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
3023 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
3024 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
3025 }
3026 on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))
3027 continuation_items = try_get(
3028 on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list)
3029 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
3030 video_items_renderer = None
3031 for key, value in continuation_item.items():
3032 if key not in known_renderers:
3033 continue
3034 video_items_renderer = {known_renderers[key][1]: continuation_items}
3035 continuation_list = [None]
3036 for entry in known_renderers[key][0](video_items_renderer):
3037 yield entry
3038 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3039 break
3040 if video_items_renderer:
3041 continue
3042 break
3043
3044 @staticmethod
3045 def _extract_selected_tab(tabs):
3046 for tab in tabs:
3047 if try_get(tab, lambda x: x['tabRenderer']['selected'], bool):
3048 return tab['tabRenderer']
3049 else:
3050 raise ExtractorError('Unable to find selected tab')
3051
3052 @staticmethod
3053 def _extract_uploader(data):
3054 uploader = {}
3055 sidebar_renderer = try_get(
3056 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3057 if sidebar_renderer:
3058 for item in sidebar_renderer:
3059 if not isinstance(item, dict):
3060 continue
3061 renderer = item.get('playlistSidebarSecondaryInfoRenderer')
3062 if not isinstance(renderer, dict):
3063 continue
3064 owner = try_get(
3065 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3066 if owner:
3067 uploader['uploader'] = owner.get('text')
3068 uploader['uploader_id'] = try_get(
3069 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3070 uploader['uploader_url'] = urljoin(
3071 'https://www.youtube.com/',
3072 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3073 return {k: v for k, v in uploader.items() if v is not None}
3074
3075 def _extract_from_tabs(self, item_id, webpage, data, tabs):
3076 playlist_id = title = description = channel_url = channel_name = channel_id = None
3077 thumbnails_list = tags = []
3078
3079 selected_tab = self._extract_selected_tab(tabs)
3080 renderer = try_get(
3081 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3082 if renderer:
3083 channel_name = renderer.get('title')
3084 channel_url = renderer.get('channelUrl')
3085 channel_id = renderer.get('externalId')
3086
3087 if not renderer:
3088 renderer = try_get(
3089 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3090 if renderer:
3091 title = renderer.get('title')
3092 description = renderer.get('description', '')
3093 playlist_id = channel_id
3094 tags = renderer.get('keywords', '').split()
3095 thumbnails_list = (
3096 try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
3097 or try_get(
3098 data,
3099 lambda x: x['sidebar']['playlistSidebarRenderer']['items'][0]['playlistSidebarPrimaryInfoRenderer']['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
3100 list)
3101 or [])
3102
3103 thumbnails = []
3104 for t in thumbnails_list:
3105 if not isinstance(t, dict):
3106 continue
3107 thumbnail_url = url_or_none(t.get('url'))
3108 if not thumbnail_url:
3109 continue
3110 thumbnails.append({
3111 'url': thumbnail_url,
3112 'width': int_or_none(t.get('width')),
3113 'height': int_or_none(t.get('height')),
3114 })
3115
3116 if playlist_id is None:
3117 playlist_id = item_id
3118 if title is None:
3119 title = playlist_id
3120 title += format_field(selected_tab, 'title', ' - %s')
3121
3122 metadata = {
3123 'playlist_id': playlist_id,
3124 'playlist_title': title,
3125 'playlist_description': description,
3126 'uploader': channel_name,
3127 'uploader_id': channel_id,
3128 'uploader_url': channel_url,
3129 'thumbnails': thumbnails,
3130 'tags': tags,
3131 }
3132 if not channel_id:
3133 metadata.update(self._extract_uploader(data))
3134 metadata.update({
3135 'channel': metadata['uploader'],
3136 'channel_id': metadata['uploader_id'],
3137 'channel_url': metadata['uploader_url']})
3138 return self.playlist_result(
3139 self._entries(
3140 selected_tab, playlist_id,
3141 self._extract_identity_token(webpage, item_id),
3142 self._extract_account_syncid(data)),
3143 **metadata)
3144
3145 def _extract_mix_playlist(self, playlist, playlist_id):
3146 first_id = last_id = None
3147 for page_num in itertools.count(1):
3148 videos = list(self._playlist_entries(playlist))
3149 if not videos:
3150 return
3151 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
3152 if start >= len(videos):
3153 return
3154 for video in videos[start:]:
3155 if video['id'] == first_id:
3156 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
3157 return
3158 yield video
3159 first_id = first_id or videos[0]['id']
3160 last_id = videos[-1]['id']
3161
3162 _, data = self._extract_webpage(
3163 'https://www.youtube.com/watch?list=%s&v=%s' % (playlist_id, last_id),
3164 '%s page %d' % (playlist_id, page_num))
3165 playlist = try_get(
3166 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3167
3168 def _extract_from_playlist(self, item_id, url, data, playlist):
3169 title = playlist.get('title') or try_get(
3170 data, lambda x: x['titleText']['simpleText'], compat_str)
3171 playlist_id = playlist.get('playlistId') or item_id
3172
3173 # Delegating everything except mix playlists to regular tab-based playlist URL
3174 playlist_url = urljoin(url, try_get(
3175 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3176 compat_str))
3177 if playlist_url and playlist_url != url:
3178 return self.url_result(
3179 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3180 video_title=title)
3181
3182 return self.playlist_result(
3183 self._extract_mix_playlist(playlist, playlist_id),
3184 playlist_id=playlist_id, playlist_title=title)
3185
3186 def _extract_alerts(self, data, expected=False):
3187
3188 def _real_extract_alerts():
3189 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
3190 if not isinstance(alert_dict, dict):
3191 continue
3192 for alert in alert_dict.values():
3193 alert_type = alert.get('type')
3194 if not alert_type:
3195 continue
3196 message = try_get(alert, lambda x: x['text']['simpleText'], compat_str)
3197 if message:
3198 yield alert_type, message
3199 for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
3200 message = try_get(run, lambda x: x['text'], compat_str)
3201 if message:
3202 yield alert_type, message
3203
3204 err_msg = None
3205 for alert_type, alert_message in _real_extract_alerts():
3206 if alert_type.lower() == 'error':
3207 if err_msg:
3208 self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
3209 err_msg = alert_message
3210 else:
3211 self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
3212
3213 if err_msg:
3214 raise ExtractorError('YouTube said: %s' % err_msg, expected=expected)
3215
3216 def _extract_webpage(self, url, item_id):
3217 retries = self._downloader.params.get('extractor_retries', 3)
3218 count = -1
3219 last_error = 'Incomplete yt initial data recieved'
3220 while count < retries:
3221 count += 1
3222 # Sometimes youtube returns a webpage with incomplete ytInitialData
3223 # See: https://github.com/yt-dlp/yt-dlp/issues/116
3224 if count:
3225 self.report_warning('%s. Retrying ...' % last_error)
3226 webpage = self._download_webpage(
3227 url, item_id,
3228 'Downloading webpage%s' % (' (retry #%d)' % count if count else ''))
3229 data = self._extract_yt_initial_data(item_id, webpage)
3230 self._extract_alerts(data, expected=True)
3231 if data.get('contents') or data.get('currentVideoEndpoint'):
3232 break
3233 if count >= retries:
3234 self._downloader.report_error(last_error)
3235 return webpage, data
3236
3237 def _real_extract(self, url):
3238 item_id = self._match_id(url)
3239 url = compat_urlparse.urlunparse(
3240 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
3241
3242 # This is not matched in a channel page with a tab selected
3243 mobj = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
3244 mobj = mobj.groupdict() if mobj else {}
3245 if mobj and not mobj.get('not_channel'):
3246 self._downloader.report_warning(
3247 'A channel/user page was given. All the channel\'s videos will be downloaded. '
3248 'To download only the videos in the home page, add a "/featured" to the URL')
3249 url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '')
3250
3251 # Handle both video/playlist URLs
3252 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3253 video_id = qs.get('v', [None])[0]
3254 playlist_id = qs.get('list', [None])[0]
3255
3256 if not video_id and (mobj.get('not_channel') or '').startswith('watch'):
3257 if not playlist_id:
3258 # If there is neither video or playlist ids,
3259 # youtube redirects to home page, which is undesirable
3260 raise ExtractorError('Unable to recognize tab page')
3261 self._downloader.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
3262 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
3263
3264 if video_id and playlist_id:
3265 if self._downloader.params.get('noplaylist'):
3266 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3267 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3268 self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
3269
3270 webpage, data = self._extract_webpage(url, item_id)
3271
3272 tabs = try_get(
3273 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
3274 if tabs:
3275 return self._extract_from_tabs(item_id, webpage, data, tabs)
3276
3277 playlist = try_get(
3278 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3279 if playlist:
3280 return self._extract_from_playlist(item_id, url, data, playlist)
3281
3282 video_id = try_get(
3283 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
3284 compat_str) or video_id
3285 if video_id:
3286 self._downloader.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
3287 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3288
3289 raise ExtractorError('Unable to recognize tab page')
3290
3291
3292class YoutubePlaylistIE(InfoExtractor):
3293 IE_DESC = 'YouTube.com playlists'
3294 _VALID_URL = r'''(?x)(?:
3295 (?:https?://)?
3296 (?:\w+\.)?
3297 (?:
3298 (?:
3299 youtube(?:kids)?\.com|
3300 invidio\.us
3301 )
3302 /.*?\?.*?\blist=
3303 )?
3304 (?P<id>%(playlist_id)s)
3305 )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3306 IE_NAME = 'youtube:playlist'
3307 _TESTS = [{
3308 'note': 'issue #673',
3309 'url': 'PLBB231211A4F62143',
3310 'info_dict': {
3311 'title': '[OLD]Team Fortress 2 (Class-based LP)',
3312 'id': 'PLBB231211A4F62143',
3313 'uploader': 'Wickydoo',
3314 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
3315 },
3316 'playlist_mincount': 29,
3317 }, {
3318 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3319 'info_dict': {
3320 'title': 'YDL_safe_search',
3321 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3322 },
3323 'playlist_count': 2,
3324 'skip': 'This playlist is private',
3325 }, {
3326 'note': 'embedded',
3327 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3328 'playlist_count': 4,
3329 'info_dict': {
3330 'title': 'JODA15',
3331 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3332 'uploader': 'milan',
3333 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
3334 }
3335 }, {
3336 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3337 'playlist_mincount': 982,
3338 'info_dict': {
3339 'title': '2018 Chinese New Singles (11/6 updated)',
3340 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3341 'uploader': 'LBK',
3342 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
3343 }
3344 }, {
3345 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
3346 'only_matching': True,
3347 }, {
3348 # music album playlist
3349 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
3350 'only_matching': True,
3351 }]
3352
3353 @classmethod
3354 def suitable(cls, url):
3355 return False if YoutubeTabIE.suitable(url) else super(
3356 YoutubePlaylistIE, cls).suitable(url)
3357
3358 def _real_extract(self, url):
3359 playlist_id = self._match_id(url)
3360 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3361 if not qs:
3362 qs = {'list': playlist_id}
3363 return self.url_result(
3364 update_url_query('https://www.youtube.com/playlist', qs),
3365 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3366
3367
3368class YoutubeYtBeIE(InfoExtractor):
3369 IE_DESC = 'youtu.be'
3370 _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}
3371 _TESTS = [{
3372 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
3373 'info_dict': {
3374 'id': 'yeWKywCrFtk',
3375 'ext': 'mp4',
3376 'title': 'Small Scale Baler and Braiding Rugs',
3377 'uploader': 'Backus-Page House Museum',
3378 'uploader_id': 'backuspagemuseum',
3379 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
3380 'upload_date': '20161008',
3381 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
3382 'categories': ['Nonprofits & Activism'],
3383 'tags': list,
3384 'like_count': int,
3385 'dislike_count': int,
3386 },
3387 'params': {
3388 'noplaylist': True,
3389 'skip_download': True,
3390 },
3391 }, {
3392 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
3393 'only_matching': True,
3394 }]
3395
3396 def _real_extract(self, url):
3397 mobj = re.match(self._VALID_URL, url)
3398 video_id = mobj.group('id')
3399 playlist_id = mobj.group('playlist_id')
3400 return self.url_result(
3401 update_url_query('https://www.youtube.com/watch', {
3402 'v': video_id,
3403 'list': playlist_id,
3404 'feature': 'youtu.be',
3405 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3406
3407
3408class YoutubeYtUserIE(InfoExtractor):
3409 IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword'
3410 _VALID_URL = r'ytuser:(?P<id>.+)'
3411 _TESTS = [{
3412 'url': 'ytuser:phihag',
3413 'only_matching': True,
3414 }]
3415
3416 def _real_extract(self, url):
3417 user_id = self._match_id(url)
3418 return self.url_result(
3419 'https://www.youtube.com/user/%s' % user_id,
3420 ie=YoutubeTabIE.ie_key(), video_id=user_id)
3421
3422
3423class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3424 IE_NAME = 'youtube:favorites'
3425 IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)'
3426 _VALID_URL = r':ytfav(?:ou?rite)?s?'
3427 _LOGIN_REQUIRED = True
3428 _TESTS = [{
3429 'url': ':ytfav',
3430 'only_matching': True,
3431 }, {
3432 'url': ':ytfavorites',
3433 'only_matching': True,
3434 }]
3435
3436 def _real_extract(self, url):
3437 return self.url_result(
3438 'https://www.youtube.com/playlist?list=LL',
3439 ie=YoutubeTabIE.ie_key())
3440
3441
3442class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
3443 IE_DESC = 'YouTube.com searches, "ytsearch" keyword'
3444 # there doesn't appear to be a real limit, for example if you search for
3445 # 'python' you get more than 8.000.000 results
3446 _MAX_RESULTS = float('inf')
3447 IE_NAME = 'youtube:search'
3448 _SEARCH_KEY = 'ytsearch'
3449 _SEARCH_PARAMS = None
3450 _TESTS = []
3451
3452 def _entries(self, query, n):
3453 data = {'query': query}
3454 if self._SEARCH_PARAMS:
3455 data['params'] = self._SEARCH_PARAMS
3456 total = 0
3457 for page_num in itertools.count(1):
3458 search = self._call_api(
3459 ep='search', video_id='query "%s"' % query, fatal=False,
3460 note='Downloading page %s' % page_num, query=data)
3461 if not search:
3462 break
3463 slr_contents = try_get(
3464 search,
3465 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
3466 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
3467 list)
3468 if not slr_contents:
3469 break
3470
3471 # Youtube sometimes adds promoted content to searches,
3472 # changing the index location of videos and token.
3473 # So we search through all entries till we find them.
3474 continuation_token = None
3475 for slr_content in slr_contents:
3476 if continuation_token is None:
3477 continuation_token = try_get(
3478 slr_content,
3479 lambda x: x['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'],
3480 compat_str)
3481
3482 isr_contents = try_get(
3483 slr_content,
3484 lambda x: x['itemSectionRenderer']['contents'],
3485 list)
3486 if not isr_contents:
3487 continue
3488 for content in isr_contents:
3489 if not isinstance(content, dict):
3490 continue
3491 video = content.get('videoRenderer')
3492 if not isinstance(video, dict):
3493 continue
3494 video_id = video.get('videoId')
3495 if not video_id:
3496 continue
3497
3498 yield self._extract_video(video)
3499 total += 1
3500 if total == n:
3501 return
3502
3503 if not continuation_token:
3504 break
3505 data['continuation'] = continuation_token
3506
3507 def _get_n_results(self, query, n):
3508 """Get a specified number of results for a query"""
3509 return self.playlist_result(self._entries(query, n), query)
3510
3511
3512class YoutubeSearchDateIE(YoutubeSearchIE):
3513 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3514 _SEARCH_KEY = 'ytsearchdate'
3515 IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword'
3516 _SEARCH_PARAMS = 'CAI%3D'
3517
3518
3519class YoutubeSearchURLIE(YoutubeSearchIE):
3520 IE_DESC = 'YouTube.com search URLs'
3521 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
3522 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
3523 # _MAX_RESULTS = 100
3524 _TESTS = [{
3525 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3526 'playlist_mincount': 5,
3527 'info_dict': {
3528 'title': 'youtube-dl test video',
3529 }
3530 }, {
3531 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3532 'only_matching': True,
3533 }]
3534
3535 @classmethod
3536 def _make_valid_url(cls):
3537 return cls._VALID_URL
3538
3539 def _real_extract(self, url):
3540 qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
3541 query = (qs.get('search_query') or qs.get('q'))[0]
3542 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
3543 return self._get_n_results(query, self._MAX_RESULTS)
3544
3545
3546class YoutubeFeedsInfoExtractor(YoutubeTabIE):
3547 """
3548 Base class for feed extractors
3549 Subclasses must define the _FEED_NAME property.
3550 """
3551 _LOGIN_REQUIRED = True
3552 _TESTS = []
3553
3554 @property
3555 def IE_NAME(self):
3556 return 'youtube:%s' % self._FEED_NAME
3557
3558 def _real_initialize(self):
3559 self._login()
3560
3561 def _real_extract(self, url):
3562 return self.url_result(
3563 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3564 ie=YoutubeTabIE.ie_key())
3565
3566
3567class YoutubeWatchLaterIE(InfoExtractor):
3568 IE_NAME = 'youtube:watchlater'
3569 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3570 _VALID_URL = r':ytwatchlater'
3571 _TESTS = [{
3572 'url': ':ytwatchlater',
3573 'only_matching': True,
3574 }]
3575
3576 def _real_extract(self, url):
3577 return self.url_result(
3578 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
3579
3580
3581class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3582 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3583 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
3584 _FEED_NAME = 'recommended'
3585 _TESTS = [{
3586 'url': ':ytrec',
3587 'only_matching': True,
3588 }, {
3589 'url': ':ytrecommended',
3590 'only_matching': True,
3591 }, {
3592 'url': 'https://youtube.com',
3593 'only_matching': True,
3594 }]
3595
3596
3597class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3598 IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)'
3599 _VALID_URL = r':ytsub(?:scription)?s?'
3600 _FEED_NAME = 'subscriptions'
3601 _TESTS = [{
3602 'url': ':ytsubs',
3603 'only_matching': True,
3604 }, {
3605 'url': ':ytsubscriptions',
3606 'only_matching': True,
3607 }]
3608
3609
3610class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3611 IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)'
3612 _VALID_URL = r':ythis(?:tory)?'
3613 _FEED_NAME = 'history'
3614 _TESTS = [{
3615 'url': ':ythistory',
3616 'only_matching': True,
3617 }]
3618
3619
3620class YoutubeTruncatedURLIE(InfoExtractor):
3621 IE_NAME = 'youtube:truncated_url'
3622 IE_DESC = False # Do not list
3623 _VALID_URL = r'''(?x)
3624 (?:https?://)?
3625 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3626 (?:watch\?(?:
3627 feature=[a-z_]+|
3628 annotation_id=annotation_[^&]+|
3629 x-yt-cl=[0-9]+|
3630 hl=[^&]*|
3631 t=[0-9]+
3632 )?
3633 |
3634 attribution_link\?a=[^&]+
3635 )
3636 $
3637 '''
3638
3639 _TESTS = [{
3640 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3641 'only_matching': True,
3642 }, {
3643 'url': 'https://www.youtube.com/watch?',
3644 'only_matching': True,
3645 }, {
3646 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3647 'only_matching': True,
3648 }, {
3649 'url': 'https://www.youtube.com/watch?feature=foo',
3650 'only_matching': True,
3651 }, {
3652 'url': 'https://www.youtube.com/watch?hl=en-GB',
3653 'only_matching': True,
3654 }, {
3655 'url': 'https://www.youtube.com/watch?t=2372',
3656 'only_matching': True,
3657 }]
3658
3659 def _real_extract(self, url):
3660 raise ExtractorError(
3661 'Did you forget to quote the URL? Remember that & is a meta '
3662 'character in most shells, so you want to put the URL in quotes, '
3663 'like youtube-dl '
3664 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3665 ' or simply youtube-dl BaW_jenozKc .',
3666 expected=True)
3667
3668
3669class YoutubeTruncatedIDIE(InfoExtractor):
3670 IE_NAME = 'youtube:truncated_id'
3671 IE_DESC = False # Do not list
3672 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3673
3674 _TESTS = [{
3675 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3676 'only_matching': True,
3677 }]
3678
3679 def _real_extract(self, url):
3680 video_id = self._match_id(url)
3681 raise ExtractorError(
3682 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3683 expected=True)