]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
projects / yt-dlp.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
[la7] Add podcasts and podcast playlists (#198)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import hashlib
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..compat import (
16 compat_chr,
17 compat_HTTPError,
18 compat_kwargs,
19 compat_parse_qs,
20 compat_str,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
24 compat_urlparse,
25 )
26 from ..jsinterp import JSInterpreter
27 from ..utils import (
28 bool_or_none,
29 clean_html,
30 dict_get,
31 ExtractorError,
32 format_field,
33 float_or_none,
34 int_or_none,
35 mimetype2ext,
36 parse_codecs,
37 parse_duration,
38 qualities,
39 remove_start,
40 smuggle_url,
41 str_or_none,
42 str_to_int,
43 try_get,
44 unescapeHTML,
45 unified_strdate,
46 unsmuggle_url,
47 update_url_query,
48 url_or_none,
49 urlencode_postdata,
50 urljoin,
51 )
52
53
54 class YoutubeBaseInfoExtractor(InfoExtractor):
55 """Provide base functions for Youtube extractors"""
56 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
57 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
58
59 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
60 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
61 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
62
63 _RESERVED_NAMES = (
64 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|'
65 r'movies|results|shared|hashtag|trending|feed|feeds|'
66 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
67
68 _NETRC_MACHINE = 'youtube'
69 # If True it will raise an error if no login info is provided
70 _LOGIN_REQUIRED = False
71
72 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
73
74 def _ids_to_results(self, ids):
75 return [
76 self.url_result(vid_id, 'Youtube', video_id=vid_id)
77 for vid_id in ids]
78
79 def _login(self):
80 """
81 Attempt to log in to YouTube.
82 True is returned if successful or skipped.
83 False is returned if login failed.
84
85 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
86 """
87 username, password = self._get_login_info()
88 # No authentication to be performed
89 if username is None:
90 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
91 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
92 # if self._downloader.params.get('cookiefile'): # TODO remove 'and False' later - too many people using outdated cookies and open issues, remind them.
93 # self.to_screen('[Cookies] Reminder - Make sure to always use up to date cookies!')
94 return True
95
96 login_page = self._download_webpage(
97 self._LOGIN_URL, None,
98 note='Downloading login page',
99 errnote='unable to fetch login page', fatal=False)
100 if login_page is False:
101 return
102
103 login_form = self._hidden_inputs(login_page)
104
105 def req(url, f_req, note, errnote):
106 data = login_form.copy()
107 data.update({
108 'pstMsg': 1,
109 'checkConnection': 'youtube',
110 'checkedDomains': 'youtube',
111 'hl': 'en',
112 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
113 'f.req': json.dumps(f_req),
114 'flowName': 'GlifWebSignIn',
115 'flowEntry': 'ServiceLogin',
116 # TODO: reverse actual botguard identifier generation algo
117 'bgRequest': '["identifier",""]',
118 })
119 return self._download_json(
120 url, None, note=note, errnote=errnote,
121 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
122 fatal=False,
123 data=urlencode_postdata(data), headers={
124 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
125 'Google-Accounts-XSRF': 1,
126 })
127
128 def warn(message):
129 self._downloader.report_warning(message)
130
131 lookup_req = [
132 username,
133 None, [], None, 'US', None, None, 2, False, True,
134 [
135 None, None,
136 [2, 1, None, 1,
137 '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',
138 None, [], 4],
139 1, [None, None, []], None, None, None, True
140 ],
141 username,
142 ]
143
144 lookup_results = req(
145 self._LOOKUP_URL, lookup_req,
146 'Looking up account info', 'Unable to look up account info')
147
148 if lookup_results is False:
149 return False
150
151 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
152 if not user_hash:
153 warn('Unable to extract user hash')
154 return False
155
156 challenge_req = [
157 user_hash,
158 None, 1, None, [1, None, None, None, [password, None, True]],
159 [
160 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],
161 1, [None, None, []], None, None, None, True
162 ]]
163
164 challenge_results = req(
165 self._CHALLENGE_URL, challenge_req,
166 'Logging in', 'Unable to log in')
167
168 if challenge_results is False:
169 return
170
171 login_res = try_get(challenge_results, lambda x: x[0][5], list)
172 if login_res:
173 login_msg = try_get(login_res, lambda x: x[5], compat_str)
174 warn(
175 'Unable to login: %s' % 'Invalid password'
176 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
177 return False
178
179 res = try_get(challenge_results, lambda x: x[0][-1], list)
180 if not res:
181 warn('Unable to extract result entry')
182 return False
183
184 login_challenge = try_get(res, lambda x: x[0][0], list)
185 if login_challenge:
186 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
187 if challenge_str == 'TWO_STEP_VERIFICATION':
188 # SEND_SUCCESS - TFA code has been successfully sent to phone
189 # QUOTA_EXCEEDED - reached the limit of TFA codes
190 status = try_get(login_challenge, lambda x: x[5], compat_str)
191 if status == 'QUOTA_EXCEEDED':
192 warn('Exceeded the limit of TFA codes, try later')
193 return False
194
195 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
196 if not tl:
197 warn('Unable to extract TL')
198 return False
199
200 tfa_code = self._get_tfa_info('2-step verification code')
201
202 if not tfa_code:
203 warn(
204 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
205 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
206 return False
207
208 tfa_code = remove_start(tfa_code, 'G-')
209
210 tfa_req = [
211 user_hash, None, 2, None,
212 [
213 9, None, None, None, None, None, None, None,
214 [None, tfa_code, True, 2]
215 ]]
216
217 tfa_results = req(
218 self._TFA_URL.format(tl), tfa_req,
219 'Submitting TFA code', 'Unable to submit TFA code')
220
221 if tfa_results is False:
222 return False
223
224 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
225 if tfa_res:
226 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
227 warn(
228 'Unable to finish TFA: %s' % 'Invalid TFA code'
229 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
230 return False
231
232 check_cookie_url = try_get(
233 tfa_results, lambda x: x[0][-1][2], compat_str)
234 else:
235 CHALLENGES = {
236 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
237 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
238 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
239 }
240 challenge = CHALLENGES.get(
241 challenge_str,
242 '%s returned error %s.' % (self.IE_NAME, challenge_str))
243 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
244 return False
245 else:
246 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
247
248 if not check_cookie_url:
249 warn('Unable to extract CheckCookie URL')
250 return False
251
252 check_cookie_results = self._download_webpage(
253 check_cookie_url, None, 'Checking cookie', fatal=False)
254
255 if check_cookie_results is False:
256 return False
257
258 if 'https://myaccount.google.com/' not in check_cookie_results:
259 warn('Unable to log in')
260 return False
261
262 return True
263
264 def _download_webpage_handle(self, *args, **kwargs):
265 query = kwargs.get('query', {}).copy()
266 kwargs['query'] = query
267 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
268 *args, **compat_kwargs(kwargs))
269
270 def _real_initialize(self):
271 if self._downloader is None:
272 return
273 if not self._login():
274 return
275
276 _YT_WEB_CLIENT_VERSION = '2.20210301.08.00'
277 _DEFAULT_API_DATA = {
278 'context': {
279 'client': {
280 'clientName': 'WEB',
281 'clientVersion': _YT_WEB_CLIENT_VERSION,
282 }
283 },
284 }
285
286 _DEFAULT_BASIC_API_HEADERS = {
287 'X-YouTube-Client-Name': '1',
288 'X-YouTube-Client-Version': _YT_WEB_CLIENT_VERSION
289 }
290
291 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
292 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
293 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
294
295 def _generate_sapisidhash_header(self):
296 sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
297 if sapisid_cookie is None:
298 return
299 time_now = round(time.time())
300 sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
301 return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
302
303 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
304 note='Downloading API JSON', errnote='Unable to download API page'):
305 data = self._DEFAULT_API_DATA.copy()
306 data.update(query)
307 headers = headers or {}
308 headers.update({'content-type': 'application/json'})
309 auth = self._generate_sapisidhash_header()
310 if auth is not None:
311 headers.update({'Authorization': auth, 'X-Origin': 'https://www.youtube.com'})
312 return self._download_json(
313 'https://www.youtube.com/youtubei/v1/%s' % ep,
314 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
315 data=json.dumps(data).encode('utf8'), headers=headers,
316 query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'})
317
318 def _extract_yt_initial_data(self, video_id, webpage):
319 return self._parse_json(
320 self._search_regex(
321 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
322 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
323 video_id)
324
325 def _extract_identity_token(self, webpage, item_id):
326 ytcfg = self._extract_ytcfg(item_id, webpage)
327 if ytcfg:
328 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
329 if token:
330 return token
331 return self._search_regex(
332 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
333 'identity token', default=None)
334
335 @staticmethod
336 def _extract_account_syncid(data):
337 """Extract syncId required to download private playlists of secondary channels"""
338 sync_ids = (
339 try_get(data, lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'], compat_str)
340 or '').split("||")
341 if len(sync_ids) >= 2 and sync_ids[1]:
342 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
343 # and just "user_syncid||" for primary channel. We only want the channel_syncid
344 return sync_ids[0]
345
346 def _extract_ytcfg(self, video_id, webpage):
347 return self._parse_json(
348 self._search_regex(
349 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
350 default='{}'), video_id, fatal=False)
351
352 def _extract_video(self, renderer):
353 video_id = renderer.get('videoId')
354 title = try_get(
355 renderer,
356 (lambda x: x['title']['runs'][0]['text'],
357 lambda x: x['title']['simpleText']), compat_str)
358 description = try_get(
359 renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'],
360 compat_str)
361 duration = parse_duration(try_get(
362 renderer, lambda x: x['lengthText']['simpleText'], compat_str))
363 view_count_text = try_get(
364 renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
365 view_count = str_to_int(self._search_regex(
366 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
367 'view count', default=None))
368 uploader = try_get(
369 renderer,
370 (lambda x: x['ownerText']['runs'][0]['text'],
371 lambda x: x['shortBylineText']['runs'][0]['text']), compat_str)
372 return {
373 '_type': 'url_transparent',
374 'ie_key': YoutubeIE.ie_key(),
375 'id': video_id,
376 'url': video_id,
377 'title': title,
378 'description': description,
379 'duration': duration,
380 'view_count': view_count,
381 'uploader': uploader,
382 }
383
384
385 class YoutubeIE(YoutubeBaseInfoExtractor):
386 IE_DESC = 'YouTube.com'
387 _INVIDIOUS_SITES = (
388 # invidious-redirect websites
389 r'(?:www\.)?redirect\.invidious\.io',
390 r'(?:(?:www|dev)\.)?invidio\.us',
391 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
392 r'(?:www\.)?invidious\.pussthecat\.org',
393 r'(?:www\.)?invidious\.048596\.xyz',
394 r'(?:www\.)?invidious\.zee\.li',
395 r'(?:www\.)?vid\.puffyan\.us',
396 r'(?:(?:www|au)\.)?ytprivate\.com',
397 r'(?:www\.)?invidious\.namazso\.eu',
398 r'(?:www\.)?invidious\.ethibox\.fr',
399 r'(?:www\.)?inv\.skyn3t\.in',
400 r'(?:www\.)?invidious\.himiko\.cloud',
401 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
402 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
403 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
404 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
405 # youtube-dl invidious instances list
406 r'(?:(?:www|no)\.)?invidiou\.sh',
407 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
408 r'(?:www\.)?invidious\.kabi\.tk',
409 r'(?:www\.)?invidious\.13ad\.de',
410 r'(?:www\.)?invidious\.mastodon\.host',
411 r'(?:www\.)?invidious\.zapashcanon\.fr',
412 r'(?:www\.)?invidious\.kavin\.rocks',
413 r'(?:www\.)?invidious\.tube',
414 r'(?:www\.)?invidiou\.site',
415 r'(?:www\.)?invidious\.site',
416 r'(?:www\.)?invidious\.xyz',
417 r'(?:www\.)?invidious\.nixnet\.xyz',
418 r'(?:www\.)?invidious\.drycat\.fr',
419 r'(?:www\.)?tube\.poal\.co',
420 r'(?:www\.)?tube\.connect\.cafe',
421 r'(?:www\.)?vid\.wxzm\.sx',
422 r'(?:www\.)?vid\.mint\.lgbt',
423 r'(?:www\.)?yewtu\.be',
424 r'(?:www\.)?yt\.elukerio\.org',
425 r'(?:www\.)?yt\.lelux\.fi',
426 r'(?:www\.)?invidious\.ggc-project\.de',
427 r'(?:www\.)?yt\.maisputain\.ovh',
428 r'(?:www\.)?invidious\.toot\.koeln',
429 r'(?:www\.)?invidious\.fdn\.fr',
430 r'(?:www\.)?watch\.nettohikari\.com',
431 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
432 r'(?:www\.)?qklhadlycap4cnod\.onion',
433 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
434 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
435 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
436 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
437 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
438 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
439 )
440 _VALID_URL = r"""(?x)^
441 (
442 (?:https?://|//) # http(s):// or protocol-independent URL
443 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
444 (?:www\.)?deturl\.com/www\.youtube\.com|
445 (?:www\.)?pwnyoutube\.com|
446 (?:www\.)?hooktube\.com|
447 (?:www\.)?yourepeat\.com|
448 tube\.majestyc\.net|
449 %(invidious)s|
450 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
451 (?:.*?\#/)? # handle anchor (#/) redirect urls
452 (?: # the various things that can precede the ID:
453 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
454 |(?: # or the v= param in all its forms
455 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
456 (?:\?|\#!?) # the params delimiter ? or # or #!
457 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
458 v=
459 )
460 ))
461 |(?:
462 youtu\.be| # just youtu.be/xxxx
463 vid\.plus| # or vid.plus/xxxx
464 zwearz\.com/watch| # or zwearz.com/watch/xxxx
465 %(invidious)s
466 )/
467 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
468 )
469 )? # all until now is optional -> you can pass the naked ID
470 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
471 (?!.*?\blist=
472 (?:
473 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
474 WL # WL are handled by the watch later IE
475 )
476 )
477 (?(1).+)? # if we found the ID, everything can follow
478 $""" % {
479 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
480 'invidious': '|'.join(_INVIDIOUS_SITES),
481 }
482 _PLAYER_INFO_RE = (
483 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
484 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
485 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
486 )
487 _formats = {
488 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
489 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
490 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
491 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
492 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
493 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
494 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
495 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
496 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
497 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
498 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
499 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
500 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
501 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
502 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
503 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
504 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
505 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
506
507
508 # 3D videos
509 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
510 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
511 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
512 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
513 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
514 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
515 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
516
517 # Apple HTTP Live Streaming
518 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
519 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
520 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
521 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
522 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
523 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
524 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
525 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
526
527 # DASH mp4 video
528 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
529 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
530 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
531 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
532 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
533 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
534 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
535 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
536 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
537 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
538 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
539 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
540
541 # Dash mp4 audio
542 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
543 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
544 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
545 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
546 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
547 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
548 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
549
550 # Dash webm
551 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
552 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
553 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
554 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
555 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
556 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
557 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
558 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
559 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
560 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
561 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
562 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
563 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
564 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
565 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
566 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
567 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
568 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
569 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
570 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
571 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
572 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
573
574 # Dash webm audio
575 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
576 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
577
578 # Dash webm audio with opus inside
579 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
580 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
581 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
582
583 # RTMP (unnamed)
584 '_rtmp': {'protocol': 'rtmp'},
585
586 # av01 video only formats sometimes served with "unknown" codecs
587 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
588 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
589 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
590 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
591 }
592 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
593
594 _GEO_BYPASS = False
595
596 IE_NAME = 'youtube'
597 _TESTS = [
598 {
599 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
600 'info_dict': {
601 'id': 'BaW_jenozKc',
602 'ext': 'mp4',
603 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
604 'uploader': 'Philipp Hagemeister',
605 'uploader_id': 'phihag',
606 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
607 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
608 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
609 'upload_date': '20121002',
610 '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 .',
611 'categories': ['Science & Technology'],
612 'tags': ['youtube-dl'],
613 'duration': 10,
614 'view_count': int,
615 'like_count': int,
616 'dislike_count': int,
617 'start_time': 1,
618 'end_time': 9,
619 }
620 },
621 {
622 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
623 'note': 'Embed-only video (#1746)',
624 'info_dict': {
625 'id': 'yZIXLfi8CZQ',
626 'ext': 'mp4',
627 'upload_date': '20120608',
628 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
629 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
630 'uploader': 'SET India',
631 'uploader_id': 'setindia',
632 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
633 'age_limit': 18,
634 },
635 'skip': 'Private video',
636 },
637 {
638 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
639 'note': 'Use the first video ID in the URL',
640 'info_dict': {
641 'id': 'BaW_jenozKc',
642 'ext': 'mp4',
643 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
644 'uploader': 'Philipp Hagemeister',
645 'uploader_id': 'phihag',
646 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
647 'upload_date': '20121002',
648 '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 .',
649 'categories': ['Science & Technology'],
650 'tags': ['youtube-dl'],
651 'duration': 10,
652 'view_count': int,
653 'like_count': int,
654 'dislike_count': int,
655 },
656 'params': {
657 'skip_download': True,
658 },
659 },
660 {
661 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
662 'note': '256k DASH audio (format 141) via DASH manifest',
663 'info_dict': {
664 'id': 'a9LDPn-MO4I',
665 'ext': 'm4a',
666 'upload_date': '20121002',
667 'uploader_id': '8KVIDEO',
668 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
669 'description': '',
670 'uploader': '8KVIDEO',
671 'title': 'UHDTV TEST 8K VIDEO.mp4'
672 },
673 'params': {
674 'youtube_include_dash_manifest': True,
675 'format': '141',
676 },
677 'skip': 'format 141 not served anymore',
678 },
679 # DASH manifest with encrypted signature
680 {
681 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
682 'info_dict': {
683 'id': 'IB3lcPjvWLA',
684 'ext': 'm4a',
685 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
686 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
687 'duration': 244,
688 'uploader': 'AfrojackVEVO',
689 'uploader_id': 'AfrojackVEVO',
690 'upload_date': '20131011',
691 'abr': 129.495,
692 },
693 'params': {
694 'youtube_include_dash_manifest': True,
695 'format': '141/bestaudio[ext=m4a]',
696 },
697 },
698 # Controversy video
699 {
700 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
701 'info_dict': {
702 'id': 'T4XJQO3qol8',
703 'ext': 'mp4',
704 'duration': 219,
705 'upload_date': '20100909',
706 'uploader': 'Amazing Atheist',
707 'uploader_id': 'TheAmazingAtheist',
708 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
709 'title': 'Burning Everyone\'s Koran',
710 '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',
711 }
712 },
713 # Normal age-gate video (embed allowed)
714 {
715 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
716 'info_dict': {
717 'id': 'HtVdAasjOgU',
718 'ext': 'mp4',
719 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
720 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
721 'duration': 142,
722 'uploader': 'The Witcher',
723 'uploader_id': 'WitcherGame',
724 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
725 'upload_date': '20140605',
726 'age_limit': 18,
727 },
728 },
729 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
730 # YouTube Red ad is not captured for creator
731 {
732 'url': '__2ABJjxzNo',
733 'info_dict': {
734 'id': '__2ABJjxzNo',
735 'ext': 'mp4',
736 'duration': 266,
737 'upload_date': '20100430',
738 'uploader_id': 'deadmau5',
739 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
740 'creator': 'deadmau5',
741 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
742 'uploader': 'deadmau5',
743 'title': 'Deadmau5 - Some Chords (HD)',
744 'alt_title': 'Some Chords',
745 },
746 'expected_warnings': [
747 'DASH manifest missing',
748 ]
749 },
750 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
751 {
752 'url': 'lqQg6PlCWgI',
753 'info_dict': {
754 'id': 'lqQg6PlCWgI',
755 'ext': 'mp4',
756 'duration': 6085,
757 'upload_date': '20150827',
758 'uploader_id': 'olympic',
759 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
760 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
761 'uploader': 'Olympic',
762 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
763 },
764 'params': {
765 'skip_download': 'requires avconv',
766 }
767 },
768 # Non-square pixels
769 {
770 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
771 'info_dict': {
772 'id': '_b-2C3KPAM0',
773 'ext': 'mp4',
774 'stretched_ratio': 16 / 9.,
775 'duration': 85,
776 'upload_date': '20110310',
777 'uploader_id': 'AllenMeow',
778 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
779 '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',
780 'uploader': '孫ᄋᄅ',
781 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
782 },
783 },
784 # url_encoded_fmt_stream_map is empty string
785 {
786 'url': 'qEJwOuvDf7I',
787 'info_dict': {
788 'id': 'qEJwOuvDf7I',
789 'ext': 'webm',
790 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
791 'description': '',
792 'upload_date': '20150404',
793 'uploader_id': 'spbelect',
794 'uploader': 'Наблюдатели Петербурга',
795 },
796 'params': {
797 'skip_download': 'requires avconv',
798 },
799 'skip': 'This live event has ended.',
800 },
801 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
802 {
803 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
804 'info_dict': {
805 'id': 'FIl7x6_3R5Y',
806 'ext': 'webm',
807 'title': 'md5:7b81415841e02ecd4313668cde88737a',
808 'description': 'md5:116377fd2963b81ec4ce64b542173306',
809 'duration': 220,
810 'upload_date': '20150625',
811 'uploader_id': 'dorappi2000',
812 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
813 'uploader': 'dorappi2000',
814 'formats': 'mincount:31',
815 },
816 'skip': 'not actual anymore',
817 },
818 # DASH manifest with segment_list
819 {
820 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
821 'md5': '8ce563a1d667b599d21064e982ab9e31',
822 'info_dict': {
823 'id': 'CsmdDsKjzN8',
824 'ext': 'mp4',
825 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
826 'uploader': 'Airtek',
827 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
828 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
829 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
830 },
831 'params': {
832 'youtube_include_dash_manifest': True,
833 'format': '135', # bestvideo
834 },
835 'skip': 'This live event has ended.',
836 },
837 {
838 # Multifeed videos (multiple cameras), URL is for Main Camera
839 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
840 'info_dict': {
841 'id': 'jvGDaLqkpTg',
842 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
843 'description': 'md5:e03b909557865076822aa169218d6a5d',
844 },
845 'playlist': [{
846 'info_dict': {
847 'id': 'jvGDaLqkpTg',
848 'ext': 'mp4',
849 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
850 'description': 'md5:e03b909557865076822aa169218d6a5d',
851 'duration': 10643,
852 'upload_date': '20161111',
853 'uploader': 'Team PGP',
854 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
855 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
856 },
857 }, {
858 'info_dict': {
859 'id': '3AKt1R1aDnw',
860 'ext': 'mp4',
861 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
862 'description': 'md5:e03b909557865076822aa169218d6a5d',
863 'duration': 10991,
864 'upload_date': '20161111',
865 'uploader': 'Team PGP',
866 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
867 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
868 },
869 }, {
870 'info_dict': {
871 'id': 'RtAMM00gpVc',
872 'ext': 'mp4',
873 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
874 'description': 'md5:e03b909557865076822aa169218d6a5d',
875 'duration': 10995,
876 'upload_date': '20161111',
877 'uploader': 'Team PGP',
878 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
879 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
880 },
881 }, {
882 'info_dict': {
883 'id': '6N2fdlP3C5U',
884 'ext': 'mp4',
885 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
886 'description': 'md5:e03b909557865076822aa169218d6a5d',
887 'duration': 10990,
888 'upload_date': '20161111',
889 'uploader': 'Team PGP',
890 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
891 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
892 },
893 }],
894 'params': {
895 'skip_download': True,
896 },
897 },
898 {
899 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
900 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
901 'info_dict': {
902 'id': 'gVfLd0zydlo',
903 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
904 },
905 'playlist_count': 2,
906 'skip': 'Not multifeed anymore',
907 },
908 {
909 'url': 'https://vid.plus/FlRa-iH7PGw',
910 'only_matching': True,
911 },
912 {
913 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
914 'only_matching': True,
915 },
916 {
917 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
918 # Also tests cut-off URL expansion in video description (see
919 # https://github.com/ytdl-org/youtube-dl/issues/1892,
920 # https://github.com/ytdl-org/youtube-dl/issues/8164)
921 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
922 'info_dict': {
923 'id': 'lsguqyKfVQg',
924 'ext': 'mp4',
925 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
926 'alt_title': 'Dark Walk - Position Music',
927 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
928 'duration': 133,
929 'upload_date': '20151119',
930 'uploader_id': 'IronSoulElf',
931 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
932 'uploader': 'IronSoulElf',
933 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
934 'track': 'Dark Walk - Position Music',
935 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
936 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
937 },
938 'params': {
939 'skip_download': True,
940 },
941 },
942 {
943 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
944 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
945 'only_matching': True,
946 },
947 {
948 # Video with yt:stretch=17:0
949 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
950 'info_dict': {
951 'id': 'Q39EVAstoRM',
952 'ext': 'mp4',
953 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
954 'description': 'md5:ee18a25c350637c8faff806845bddee9',
955 'upload_date': '20151107',
956 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
957 'uploader': 'CH GAMER DROID',
958 },
959 'params': {
960 'skip_download': True,
961 },
962 'skip': 'This video does not exist.',
963 },
964 {
965 # Video licensed under Creative Commons
966 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
967 'info_dict': {
968 'id': 'M4gD1WSo5mA',
969 'ext': 'mp4',
970 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
971 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
972 'duration': 721,
973 'upload_date': '20150127',
974 'uploader_id': 'BerkmanCenter',
975 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
976 'uploader': 'The Berkman Klein Center for Internet & Society',
977 'license': 'Creative Commons Attribution license (reuse allowed)',
978 },
979 'params': {
980 'skip_download': True,
981 },
982 },
983 {
984 # Channel-like uploader_url
985 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
986 'info_dict': {
987 'id': 'eQcmzGIKrzg',
988 'ext': 'mp4',
989 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
990 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
991 'duration': 4060,
992 'upload_date': '20151119',
993 'uploader': 'Bernie Sanders',
994 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
995 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
996 'license': 'Creative Commons Attribution license (reuse allowed)',
997 },
998 'params': {
999 'skip_download': True,
1000 },
1001 },
1002 {
1003 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1004 'only_matching': True,
1005 },
1006 {
1007 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1008 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1009 'only_matching': True,
1010 },
1011 {
1012 # Rental video preview
1013 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1014 'info_dict': {
1015 'id': 'uGpuVWrhIzE',
1016 'ext': 'mp4',
1017 'title': 'Piku - Trailer',
1018 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1019 'upload_date': '20150811',
1020 'uploader': 'FlixMatrix',
1021 'uploader_id': 'FlixMatrixKaravan',
1022 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1023 'license': 'Standard YouTube License',
1024 },
1025 'params': {
1026 'skip_download': True,
1027 },
1028 'skip': 'This video is not available.',
1029 },
1030 {
1031 # YouTube Red video with episode data
1032 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1033 'info_dict': {
1034 'id': 'iqKdEhx-dD4',
1035 'ext': 'mp4',
1036 'title': 'Isolation - Mind Field (Ep 1)',
1037 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1038 'duration': 2085,
1039 'upload_date': '20170118',
1040 'uploader': 'Vsauce',
1041 'uploader_id': 'Vsauce',
1042 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1043 'series': 'Mind Field',
1044 'season_number': 1,
1045 'episode_number': 1,
1046 },
1047 'params': {
1048 'skip_download': True,
1049 },
1050 'expected_warnings': [
1051 'Skipping DASH manifest',
1052 ],
1053 },
1054 {
1055 # The following content has been identified by the YouTube community
1056 # as inappropriate or offensive to some audiences.
1057 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1058 'info_dict': {
1059 'id': '6SJNVb0GnPI',
1060 'ext': 'mp4',
1061 'title': 'Race Differences in Intelligence',
1062 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1063 'duration': 965,
1064 'upload_date': '20140124',
1065 'uploader': 'New Century Foundation',
1066 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1067 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1068 },
1069 'params': {
1070 'skip_download': True,
1071 },
1072 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1073 },
1074 {
1075 # itag 212
1076 'url': '1t24XAntNCY',
1077 'only_matching': True,
1078 },
1079 {
1080 # geo restricted to JP
1081 'url': 'sJL6WA-aGkQ',
1082 'only_matching': True,
1083 },
1084 {
1085 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1086 'only_matching': True,
1087 },
1088 {
1089 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1090 'only_matching': True,
1091 },
1092 {
1093 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1094 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1095 'only_matching': True,
1096 },
1097 {
1098 # DRM protected
1099 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1100 'only_matching': True,
1101 },
1102 {
1103 # Video with unsupported adaptive stream type formats
1104 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1105 'info_dict': {
1106 'id': 'Z4Vy8R84T1U',
1107 'ext': 'mp4',
1108 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1109 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1110 'duration': 433,
1111 'upload_date': '20130923',
1112 'uploader': 'Amelia Putri Harwita',
1113 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1114 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1115 'formats': 'maxcount:10',
1116 },
1117 'params': {
1118 'skip_download': True,
1119 'youtube_include_dash_manifest': False,
1120 },
1121 'skip': 'not actual anymore',
1122 },
1123 {
1124 # Youtube Music Auto-generated description
1125 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1126 'info_dict': {
1127 'id': 'MgNrAu2pzNs',
1128 'ext': 'mp4',
1129 'title': 'Voyeur Girl',
1130 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1131 'upload_date': '20190312',
1132 'uploader': 'Stephen - Topic',
1133 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1134 'artist': 'Stephen',
1135 'track': 'Voyeur Girl',
1136 'album': 'it\'s too much love to know my dear',
1137 'release_date': '20190313',
1138 'release_year': 2019,
1139 },
1140 'params': {
1141 'skip_download': True,
1142 },
1143 },
1144 {
1145 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1146 'only_matching': True,
1147 },
1148 {
1149 # invalid -> valid video id redirection
1150 'url': 'DJztXj2GPfl',
1151 'info_dict': {
1152 'id': 'DJztXj2GPfk',
1153 'ext': 'mp4',
1154 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1155 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1156 'upload_date': '20090125',
1157 'uploader': 'Prochorowka',
1158 'uploader_id': 'Prochorowka',
1159 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1160 'artist': 'Panjabi MC',
1161 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1162 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1163 },
1164 'params': {
1165 'skip_download': True,
1166 },
1167 'skip': 'Video unavailable',
1168 },
1169 {
1170 # empty description results in an empty string
1171 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1172 'info_dict': {
1173 'id': 'x41yOUIvK2k',
1174 'ext': 'mp4',
1175 'title': 'IMG 3456',
1176 'description': '',
1177 'upload_date': '20170613',
1178 'uploader_id': 'ElevageOrVert',
1179 'uploader': 'ElevageOrVert',
1180 },
1181 'params': {
1182 'skip_download': True,
1183 },
1184 },
1185 {
1186 # with '};' inside yt initial data (see [1])
1187 # see [2] for an example with '};' inside ytInitialPlayerResponse
1188 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1189 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1190 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1191 'info_dict': {
1192 'id': 'CHqg6qOn4no',
1193 'ext': 'mp4',
1194 'title': 'Part 77 Sort a list of simple types in c#',
1195 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1196 'upload_date': '20130831',
1197 'uploader_id': 'kudvenkat',
1198 'uploader': 'kudvenkat',
1199 },
1200 'params': {
1201 'skip_download': True,
1202 },
1203 },
1204 {
1205 # another example of '};' in ytInitialData
1206 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1207 'only_matching': True,
1208 },
1209 {
1210 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1211 'only_matching': True,
1212 },
1213 {
1214 # https://github.com/ytdl-org/youtube-dl/pull/28094
1215 'url': 'OtqTfy26tG0',
1216 'info_dict': {
1217 'id': 'OtqTfy26tG0',
1218 'ext': 'mp4',
1219 'title': 'Burn Out',
1220 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1221 'upload_date': '20141120',
1222 'uploader': 'The Cinematic Orchestra - Topic',
1223 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1224 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1225 'artist': 'The Cinematic Orchestra',
1226 'track': 'Burn Out',
1227 'album': 'Every Day',
1228 'release_data': None,
1229 'release_year': None,
1230 },
1231 'params': {
1232 'skip_download': True,
1233 },
1234 },
1235 {
1236 # controversial video, only works with bpctr when authenticated with cookies
1237 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1238 'only_matching': True,
1239 },
1240 ]
1241
1242 def __init__(self, *args, **kwargs):
1243 super(YoutubeIE, self).__init__(*args, **kwargs)
1244 self._code_cache = {}
1245 self._player_cache = {}
1246
1247 def _signature_cache_id(self, example_sig):
1248 """ Return a string representation of a signature """
1249 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1250
1251 @classmethod
1252 def _extract_player_info(cls, player_url):
1253 for player_re in cls._PLAYER_INFO_RE:
1254 id_m = re.search(player_re, player_url)
1255 if id_m:
1256 break
1257 else:
1258 raise ExtractorError('Cannot identify player %r' % player_url)
1259 return id_m.group('id')
1260
1261 def _extract_signature_function(self, video_id, player_url, example_sig):
1262 player_id = self._extract_player_info(player_url)
1263
1264 # Read from filesystem cache
1265 func_id = 'js_%s_%s' % (
1266 player_id, self._signature_cache_id(example_sig))
1267 assert os.path.basename(func_id) == func_id
1268
1269 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1270 if cache_spec is not None:
1271 return lambda s: ''.join(s[i] for i in cache_spec)
1272
1273 if player_id not in self._code_cache:
1274 self._code_cache[player_id] = self._download_webpage(
1275 player_url, video_id,
1276 note='Downloading player ' + player_id,
1277 errnote='Download of %s failed' % player_url)
1278 code = self._code_cache[player_id]
1279 res = self._parse_sig_js(code)
1280
1281 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1282 cache_res = res(test_string)
1283 cache_spec = [ord(c) for c in cache_res]
1284
1285 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1286 return res
1287
1288 def _print_sig_code(self, func, example_sig):
1289 def gen_sig_code(idxs):
1290 def _genslice(start, end, step):
1291 starts = '' if start == 0 else str(start)
1292 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1293 steps = '' if step == 1 else (':%d' % step)
1294 return 's[%s%s%s]' % (starts, ends, steps)
1295
1296 step = None
1297 # Quelch pyflakes warnings - start will be set when step is set
1298 start = '(Never used)'
1299 for i, prev in zip(idxs[1:], idxs[:-1]):
1300 if step is not None:
1301 if i - prev == step:
1302 continue
1303 yield _genslice(start, prev, step)
1304 step = None
1305 continue
1306 if i - prev in [-1, 1]:
1307 step = i - prev
1308 start = prev
1309 continue
1310 else:
1311 yield 's[%d]' % prev
1312 if step is None:
1313 yield 's[%d]' % i
1314 else:
1315 yield _genslice(start, i, step)
1316
1317 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1318 cache_res = func(test_string)
1319 cache_spec = [ord(c) for c in cache_res]
1320 expr_code = ' + '.join(gen_sig_code(cache_spec))
1321 signature_id_tuple = '(%s)' % (
1322 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1323 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1324 ' return %s\n') % (signature_id_tuple, expr_code)
1325 self.to_screen('Extracted signature function:\n' + code)
1326
1327 def _parse_sig_js(self, jscode):
1328 funcname = self._search_regex(
1329 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1330 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1331 r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
1332 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
1333 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}\(a,\d+\)',
1334 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\(\s*""\s*\)',
1335 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1336 # Obsolete patterns
1337 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1338 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1339 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1340 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1341 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1342 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1343 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1344 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1345 jscode, 'Initial JS player signature function name', group='sig')
1346
1347 jsi = JSInterpreter(jscode)
1348 initial_function = jsi.extract_function(funcname)
1349 return lambda s: initial_function([s])
1350
1351 def _decrypt_signature(self, s, video_id, player_url):
1352 """Turn the encrypted s field into a working signature"""
1353
1354 if player_url is None:
1355 raise ExtractorError('Cannot decrypt signature without player_url')
1356
1357 if player_url.startswith('//'):
1358 player_url = 'https:' + player_url
1359 elif not re.match(r'https?://', player_url):
1360 player_url = compat_urlparse.urljoin(
1361 'https://www.youtube.com', player_url)
1362 try:
1363 player_id = (player_url, self._signature_cache_id(s))
1364 if player_id not in self._player_cache:
1365 func = self._extract_signature_function(
1366 video_id, player_url, s
1367 )
1368 self._player_cache[player_id] = func
1369 func = self._player_cache[player_id]
1370 if self._downloader.params.get('youtube_print_sig_code'):
1371 self._print_sig_code(func, s)
1372 return func(s)
1373 except Exception as e:
1374 tb = traceback.format_exc()
1375 raise ExtractorError(
1376 'Signature extraction failed: ' + tb, cause=e)
1377
1378 def _mark_watched(self, video_id, player_response):
1379 playback_url = url_or_none(try_get(
1380 player_response,
1381 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']))
1382 if not playback_url:
1383 return
1384 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1385 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1386
1387 # cpn generation algorithm is reverse engineered from base.js.
1388 # In fact it works even with dummy cpn.
1389 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1390 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1391
1392 qs.update({
1393 'ver': ['2'],
1394 'cpn': [cpn],
1395 })
1396 playback_url = compat_urlparse.urlunparse(
1397 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1398
1399 self._download_webpage(
1400 playback_url, video_id, 'Marking watched',
1401 'Unable to mark watched', fatal=False)
1402
1403 @staticmethod
1404 def _extract_urls(webpage):
1405 # Embedded YouTube player
1406 entries = [
1407 unescapeHTML(mobj.group('url'))
1408 for mobj in re.finditer(r'''(?x)
1409 (?:
1410 <iframe[^>]+?src=|
1411 data-video-url=|
1412 <embed[^>]+?src=|
1413 embedSWF\(?:\s*|
1414 <object[^>]+data=|
1415 new\s+SWFObject\(
1416 )
1417 (["\'])
1418 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1419 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1420 \1''', webpage)]
1421
1422 # lazyYT YouTube embed
1423 entries.extend(list(map(
1424 unescapeHTML,
1425 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1426
1427 # Wordpress "YouTube Video Importer" plugin
1428 matches = re.findall(r'''(?x)<div[^>]+
1429 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1430 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1431 entries.extend(m[-1] for m in matches)
1432
1433 return entries
1434
1435 @staticmethod
1436 def _extract_url(webpage):
1437 urls = YoutubeIE._extract_urls(webpage)
1438 return urls[0] if urls else None
1439
1440 @classmethod
1441 def extract_id(cls, url):
1442 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1443 if mobj is None:
1444 raise ExtractorError('Invalid URL: %s' % url)
1445 video_id = mobj.group(2)
1446 return video_id
1447
1448 def _extract_chapters_from_json(self, data, video_id, duration):
1449 chapters_list = try_get(
1450 data,
1451 lambda x: x['playerOverlays']
1452 ['playerOverlayRenderer']
1453 ['decoratedPlayerBarRenderer']
1454 ['decoratedPlayerBarRenderer']
1455 ['playerBar']
1456 ['chapteredPlayerBarRenderer']
1457 ['chapters'],
1458 list)
1459 if not chapters_list:
1460 return
1461
1462 def chapter_time(chapter):
1463 return float_or_none(
1464 try_get(
1465 chapter,
1466 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1467 int),
1468 scale=1000)
1469 chapters = []
1470 for next_num, chapter in enumerate(chapters_list, start=1):
1471 start_time = chapter_time(chapter)
1472 if start_time is None:
1473 continue
1474 end_time = (chapter_time(chapters_list[next_num])
1475 if next_num < len(chapters_list) else duration)
1476 if end_time is None:
1477 continue
1478 title = try_get(
1479 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1480 compat_str)
1481 chapters.append({
1482 'start_time': start_time,
1483 'end_time': end_time,
1484 'title': title,
1485 })
1486 return chapters
1487
1488 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
1489 return self._parse_json(self._search_regex(
1490 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
1491 regex), webpage, name, default='{}'), video_id, fatal=False)
1492
1493 @staticmethod
1494 def _join_text_entries(runs):
1495 text = None
1496 for run in runs:
1497 if not isinstance(run, dict):
1498 continue
1499 sub_text = try_get(run, lambda x: x['text'], compat_str)
1500 if sub_text:
1501 if not text:
1502 text = sub_text
1503 continue
1504 text += sub_text
1505 return text
1506
1507 def _extract_comment(self, comment_renderer, parent=None):
1508 comment_id = comment_renderer.get('commentId')
1509 if not comment_id:
1510 return
1511 comment_text_runs = try_get(comment_renderer, lambda x: x['contentText']['runs']) or []
1512 text = self._join_text_entries(comment_text_runs) or ''
1513 comment_time_text = try_get(comment_renderer, lambda x: x['publishedTimeText']['runs']) or []
1514 time_text = self._join_text_entries(comment_time_text)
1515
1516 author = try_get(comment_renderer, lambda x: x['authorText']['simpleText'], compat_str)
1517 author_id = try_get(comment_renderer,
1518 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
1519 votes = str_to_int(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
1520 lambda x: x['likeCount']), compat_str)) or 0
1521 author_thumbnail = try_get(comment_renderer,
1522 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
1523
1524 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
1525 is_liked = try_get(comment_renderer, lambda x: x['isLiked'], bool)
1526
1527 return {
1528 'id': comment_id,
1529 'text': text,
1530 # TODO: This should be parsed to timestamp
1531 'time_text': time_text,
1532 'like_count': votes,
1533 'is_favorited': is_liked,
1534 'author': author,
1535 'author_id': author_id,
1536 'author_thumbnail': author_thumbnail,
1537 'author_is_uploader': author_is_uploader,
1538 'parent': parent or 'root'
1539 }
1540
1541 def _comment_entries(self, root_continuation_data, identity_token, account_syncid,
1542 session_token_list, parent=None, comment_counts=None):
1543
1544 def extract_thread(parent_renderer):
1545 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
1546 if not parent:
1547 comment_counts[2] = 0
1548 for content in contents:
1549 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
1550 comment_renderer = try_get(
1551 comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
1552 content, (lambda x: x['commentRenderer'], dict))
1553
1554 if not comment_renderer:
1555 continue
1556 comment = self._extract_comment(comment_renderer, parent)
1557 if not comment:
1558 continue
1559 comment_counts[0] += 1
1560 yield comment
1561 # Attempt to get the replies
1562 comment_replies_renderer = try_get(
1563 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
1564
1565 if comment_replies_renderer:
1566 comment_counts[2] += 1
1567 comment_entries_iter = self._comment_entries(
1568 comment_replies_renderer, identity_token, account_syncid,
1569 parent=comment.get('id'), session_token_list=session_token_list,
1570 comment_counts=comment_counts)
1571
1572 for reply_comment in comment_entries_iter:
1573 yield reply_comment
1574
1575 if not comment_counts:
1576 # comment so far, est. total comments, current comment thread #
1577 comment_counts = [0, 0, 0]
1578 headers = self._DEFAULT_BASIC_API_HEADERS.copy()
1579
1580 # TODO: Generalize the download code with TabIE
1581 if identity_token:
1582 headers['x-youtube-identity-token'] = identity_token
1583
1584 if account_syncid:
1585 headers['X-Goog-PageId'] = account_syncid
1586 headers['X-Goog-AuthUser'] = 0
1587
1588 continuation = YoutubeTabIE._extract_continuation(root_continuation_data) # TODO
1589 first_continuation = False
1590 if parent is None:
1591 first_continuation = True
1592
1593 for page_num in itertools.count(0):
1594 if not continuation:
1595 break
1596 retries = self._downloader.params.get('extractor_retries', 3)
1597 count = -1
1598 last_error = None
1599
1600 while count < retries:
1601 count += 1
1602 if last_error:
1603 self.report_warning('%s. Retrying ...' % last_error)
1604 try:
1605 query = {
1606 'ctoken': continuation['ctoken'],
1607 'pbj': 1,
1608 'type': 'next',
1609 }
1610 if parent:
1611 query['action_get_comment_replies'] = 1
1612 else:
1613 query['action_get_comments'] = 1
1614
1615 comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
1616 if page_num == 0:
1617 if first_continuation:
1618 note_prefix = "Downloading initial comment continuation page"
1619 else:
1620 note_prefix = " Downloading comment reply thread %d %s" % (comment_counts[2], comment_prog_str)
1621 else:
1622 note_prefix = "%sDownloading comment%s page %d %s" % (
1623 " " if parent else "",
1624 ' replies' if parent else '',
1625 page_num,
1626 comment_prog_str)
1627
1628 browse = self._download_json(
1629 'https://www.youtube.com/comment_service_ajax', None,
1630 '%s %s' % (note_prefix, '(retry #%d)' % count if count else ''),
1631 headers=headers, query=query,
1632 data=urlencode_postdata({
1633 'session_token': session_token_list[0]
1634 }))
1635 except ExtractorError as e:
1636 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404, 413):
1637 if e.cause.code == 413:
1638 self.report_warning("Assumed end of comments (received HTTP Error 413)")
1639 return
1640 # Downloading page may result in intermittent 5xx HTTP error
1641 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
1642 last_error = 'HTTP Error %s' % e.cause.code
1643 if e.cause.code == 404:
1644 last_error = last_error + " (this API is probably deprecated)"
1645 if count < retries:
1646 continue
1647 raise
1648 else:
1649 session_token = try_get(browse, lambda x: x['xsrf_token'], compat_str)
1650 if session_token:
1651 session_token_list[0] = session_token
1652
1653 response = try_get(browse,
1654 (lambda x: x['response'],
1655 lambda x: x[1]['response'])) or {}
1656
1657 if response.get('continuationContents'):
1658 break
1659
1660 # YouTube sometimes gives reload: now json if something went wrong (e.g. bad auth)
1661 if browse.get('reload'):
1662 raise ExtractorError("Invalid or missing params in continuation request", expected=False)
1663
1664 # TODO: not tested, merged from old extractor
1665 err_msg = browse.get('externalErrorMessage')
1666 if err_msg:
1667 raise ExtractorError('YouTube said: %s' % err_msg, expected=False)
1668
1669 # Youtube sometimes sends incomplete data
1670 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1671 last_error = 'Incomplete data received'
1672 if count >= retries:
1673 self._downloader.report_error(last_error)
1674
1675 if not response:
1676 break
1677
1678 known_continuation_renderers = {
1679 'itemSectionContinuation': extract_thread,
1680 'commentRepliesContinuation': extract_thread
1681 }
1682
1683 # extract next root continuation from the results
1684 continuation_contents = try_get(
1685 response, lambda x: x['continuationContents'], dict) or {}
1686
1687 for key, value in continuation_contents.items():
1688 if key not in known_continuation_renderers:
1689 continue
1690 continuation_renderer = value
1691
1692 if first_continuation:
1693 first_continuation = False
1694 expected_comment_count = try_get(
1695 continuation_renderer,
1696 (lambda x: x['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'],
1697 lambda x: x['header']['commentsHeaderRenderer']['commentsCount']['runs'][0]['text']),
1698 compat_str)
1699
1700 if expected_comment_count:
1701 comment_counts[1] = str_to_int(expected_comment_count)
1702 self.to_screen("Downloading ~%d comments" % str_to_int(expected_comment_count))
1703 yield comment_counts[1]
1704
1705 # TODO: cli arg.
1706 # 1/True for newest, 0/False for popular (default)
1707 comment_sort_index = int(True)
1708 sort_continuation_renderer = try_get(
1709 continuation_renderer,
1710 lambda x: x['header']['commentsHeaderRenderer']['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems']
1711 [comment_sort_index]['continuation']['reloadContinuationData'], dict)
1712 # If this fails, the initial continuation page
1713 # starts off with popular anyways.
1714 if sort_continuation_renderer:
1715 continuation = YoutubeTabIE._build_continuation_query(
1716 continuation=sort_continuation_renderer.get('continuation'),
1717 ctp=sort_continuation_renderer.get('clickTrackingParams'))
1718 self.to_screen("Sorting comments by %s" % ('popular' if comment_sort_index == 0 else 'newest'))
1719 break
1720
1721 for entry in known_continuation_renderers[key](continuation_renderer):
1722 yield entry
1723
1724 continuation = YoutubeTabIE._extract_continuation(continuation_renderer) # TODO
1725 break
1726
1727 def _extract_comments(self, ytcfg, video_id, contents, webpage, xsrf_token):
1728 """Entry for comment extraction"""
1729 comments = []
1730 known_entry_comment_renderers = (
1731 'itemSectionRenderer',
1732 )
1733 estimated_total = 0
1734 for entry in contents:
1735 for key, renderer in entry.items():
1736 if key not in known_entry_comment_renderers:
1737 continue
1738
1739 comment_iter = self._comment_entries(
1740 renderer,
1741 identity_token=self._extract_identity_token(webpage, item_id=video_id),
1742 account_syncid=self._extract_account_syncid(ytcfg),
1743 session_token_list=[xsrf_token])
1744
1745 for comment in comment_iter:
1746 if isinstance(comment, int):
1747 estimated_total = comment
1748 continue
1749 comments.append(comment)
1750 break
1751 self.to_screen("Downloaded %d/%d comments" % (len(comments), estimated_total))
1752 return {
1753 'comments': comments,
1754 'comment_count': len(comments),
1755 }
1756
1757 def _real_extract(self, url):
1758 url, smuggled_data = unsmuggle_url(url, {})
1759 video_id = self._match_id(url)
1760 base_url = self.http_scheme() + '//www.youtube.com/'
1761 webpage_url = base_url + 'watch?v=' + video_id
1762 webpage = self._download_webpage(
1763 webpage_url + '&has_verified=1&bpctr=9999999999',
1764 video_id, fatal=False)
1765
1766 player_response = None
1767 if webpage:
1768 player_response = self._extract_yt_initial_variable(
1769 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
1770 video_id, 'initial player response')
1771 if not player_response:
1772 player_response = self._call_api(
1773 'player', {'videoId': video_id}, video_id)
1774
1775 playability_status = player_response.get('playabilityStatus') or {}
1776 if playability_status.get('reason') == 'Sign in to confirm your age':
1777 pr = self._parse_json(try_get(compat_parse_qs(
1778 self._download_webpage(
1779 base_url + 'get_video_info', video_id,
1780 'Refetching age-gated info webpage',
1781 'unable to download video info webpage', query={
1782 'video_id': video_id,
1783 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1784 }, fatal=False)),
1785 lambda x: x['player_response'][0],
1786 compat_str) or '{}', video_id)
1787 if pr:
1788 player_response = pr
1789
1790 trailer_video_id = try_get(
1791 playability_status,
1792 lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'],
1793 compat_str)
1794 if trailer_video_id:
1795 return self.url_result(
1796 trailer_video_id, self.ie_key(), trailer_video_id)
1797
1798 def get_text(x):
1799 if not x:
1800 return
1801 return x.get('simpleText') or ''.join([r['text'] for r in x['runs']])
1802
1803 search_meta = (
1804 lambda x: self._html_search_meta(x, webpage, default=None)) \
1805 if webpage else lambda x: None
1806
1807 video_details = player_response.get('videoDetails') or {}
1808 microformat = try_get(
1809 player_response,
1810 lambda x: x['microformat']['playerMicroformatRenderer'],
1811 dict) or {}
1812 video_title = video_details.get('title') \
1813 or get_text(microformat.get('title')) \
1814 or search_meta(['og:title', 'twitter:title', 'title'])
1815 video_description = video_details.get('shortDescription')
1816
1817 if not smuggled_data.get('force_singlefeed', False):
1818 if not self._downloader.params.get('noplaylist'):
1819 multifeed_metadata_list = try_get(
1820 player_response,
1821 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1822 compat_str)
1823 if multifeed_metadata_list:
1824 entries = []
1825 feed_ids = []
1826 for feed in multifeed_metadata_list.split(','):
1827 # Unquote should take place before split on comma (,) since textual
1828 # fields may contain comma as well (see
1829 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1830 feed_data = compat_parse_qs(
1831 compat_urllib_parse_unquote_plus(feed))
1832
1833 def feed_entry(name):
1834 return try_get(
1835 feed_data, lambda x: x[name][0], compat_str)
1836
1837 feed_id = feed_entry('id')
1838 if not feed_id:
1839 continue
1840 feed_title = feed_entry('title')
1841 title = video_title
1842 if feed_title:
1843 title += ' (%s)' % feed_title
1844 entries.append({
1845 '_type': 'url_transparent',
1846 'ie_key': 'Youtube',
1847 'url': smuggle_url(
1848 base_url + 'watch?v=' + feed_data['id'][0],
1849 {'force_singlefeed': True}),
1850 'title': title,
1851 })
1852 feed_ids.append(feed_id)
1853 self.to_screen(
1854 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1855 % (', '.join(feed_ids), video_id))
1856 return self.playlist_result(
1857 entries, video_id, video_title, video_description)
1858 else:
1859 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1860
1861 formats = []
1862 itags = []
1863 itag_qualities = {}
1864 player_url = None
1865 q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'])
1866 streaming_data = player_response.get('streamingData') or {}
1867 streaming_formats = streaming_data.get('formats') or []
1868 streaming_formats.extend(streaming_data.get('adaptiveFormats') or [])
1869 for fmt in streaming_formats:
1870 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
1871 continue
1872
1873 itag = str_or_none(fmt.get('itag'))
1874 quality = fmt.get('quality')
1875 if itag and quality:
1876 itag_qualities[itag] = quality
1877 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
1878 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
1879 # number of fragment that would subsequently requested with (`&sq=N`)
1880 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
1881 continue
1882
1883 fmt_url = fmt.get('url')
1884 if not fmt_url:
1885 sc = compat_parse_qs(fmt.get('signatureCipher'))
1886 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
1887 encrypted_sig = try_get(sc, lambda x: x['s'][0])
1888 if not (sc and fmt_url and encrypted_sig):
1889 continue
1890 if not player_url:
1891 if not webpage:
1892 continue
1893 player_url = self._search_regex(
1894 r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"',
1895 webpage, 'player URL', fatal=False)
1896 if not player_url:
1897 continue
1898 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
1899 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
1900 fmt_url += '&' + sp + '=' + signature
1901
1902 if itag:
1903 itags.append(itag)
1904 tbr = float_or_none(
1905 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
1906 dct = {
1907 'asr': int_or_none(fmt.get('audioSampleRate')),
1908 'filesize': int_or_none(fmt.get('contentLength')),
1909 'format_id': itag,
1910 'format_note': fmt.get('qualityLabel') or quality,
1911 'fps': int_or_none(fmt.get('fps')),
1912 'height': int_or_none(fmt.get('height')),
1913 'quality': q(quality),
1914 'tbr': tbr,
1915 'url': fmt_url,
1916 'width': fmt.get('width'),
1917 }
1918 mimetype = fmt.get('mimeType')
1919 if mimetype:
1920 mobj = re.match(
1921 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype)
1922 if mobj:
1923 dct['ext'] = mimetype2ext(mobj.group(1))
1924 dct.update(parse_codecs(mobj.group(2)))
1925 no_audio = dct.get('acodec') == 'none'
1926 no_video = dct.get('vcodec') == 'none'
1927 if no_audio:
1928 dct['vbr'] = tbr
1929 if no_video:
1930 dct['abr'] = tbr
1931 if no_audio or no_video:
1932 dct['downloader_options'] = {
1933 # Youtube throttles chunks >~10M
1934 'http_chunk_size': 10485760,
1935 }
1936 if dct.get('ext'):
1937 dct['container'] = dct['ext'] + '_dash'
1938 formats.append(dct)
1939
1940 hls_manifest_url = streaming_data.get('hlsManifestUrl')
1941 if hls_manifest_url:
1942 for f in self._extract_m3u8_formats(
1943 hls_manifest_url, video_id, 'mp4', fatal=False):
1944 itag = self._search_regex(
1945 r'/itag/(\d+)', f['url'], 'itag', default=None)
1946 if itag:
1947 f['format_id'] = itag
1948 formats.append(f)
1949
1950 if self._downloader.params.get('youtube_include_dash_manifest', True):
1951 dash_manifest_url = streaming_data.get('dashManifestUrl')
1952 if dash_manifest_url:
1953 for f in self._extract_mpd_formats(
1954 dash_manifest_url, video_id, fatal=False):
1955 itag = f['format_id']
1956 if itag in itags:
1957 continue
1958 if itag in itag_qualities:
1959 # Not actually usefull since the sorting is already done with "quality,res,fps,codec"
1960 # but kept to maintain feature parity (and code similarity) with youtube-dl
1961 # Remove if this causes any issues with sorting in future
1962 f['quality'] = q(itag_qualities[itag])
1963 filesize = int_or_none(self._search_regex(
1964 r'/clen/(\d+)', f.get('fragment_base_url')
1965 or f['url'], 'file size', default=None))
1966 if filesize:
1967 f['filesize'] = filesize
1968 formats.append(f)
1969
1970 if not formats:
1971 if not self._downloader.params.get('allow_unplayable_formats') and streaming_data.get('licenseInfos'):
1972 raise ExtractorError(
1973 'This video is DRM protected.', expected=True)
1974 pemr = try_get(
1975 playability_status,
1976 lambda x: x['errorScreen']['playerErrorMessageRenderer'],
1977 dict) or {}
1978 reason = get_text(pemr.get('reason')) or playability_status.get('reason')
1979 subreason = pemr.get('subreason')
1980 if subreason:
1981 subreason = clean_html(get_text(subreason))
1982 if subreason == 'The uploader has not made this video available in your country.':
1983 countries = microformat.get('availableCountries')
1984 if not countries:
1985 regions_allowed = search_meta('regionsAllowed')
1986 countries = regions_allowed.split(',') if regions_allowed else None
1987 self.raise_geo_restricted(
1988 subreason, countries)
1989 reason += '\n' + subreason
1990 if reason:
1991 raise ExtractorError(reason, expected=True)
1992
1993 self._sort_formats(formats)
1994
1995 keywords = video_details.get('keywords') or []
1996 if not keywords and webpage:
1997 keywords = [
1998 unescapeHTML(m.group('content'))
1999 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
2000 for keyword in keywords:
2001 if keyword.startswith('yt:stretch='):
2002 w, h = keyword.split('=')[1].split(':')
2003 w, h = int(w), int(h)
2004 if w > 0 and h > 0:
2005 ratio = w / h
2006 for f in formats:
2007 if f.get('vcodec') != 'none':
2008 f['stretched_ratio'] = ratio
2009
2010 thumbnails = []
2011 for container in (video_details, microformat):
2012 for thumbnail in (try_get(
2013 container,
2014 lambda x: x['thumbnail']['thumbnails'], list) or []):
2015 thumbnail_url = thumbnail.get('url')
2016 if not thumbnail_url:
2017 continue
2018 thumbnails.append({
2019 'height': int_or_none(thumbnail.get('height')),
2020 'url': thumbnail_url,
2021 'width': int_or_none(thumbnail.get('width')),
2022 })
2023 if thumbnails:
2024 break
2025 else:
2026 thumbnail = search_meta(['og:image', 'twitter:image'])
2027 if thumbnail:
2028 thumbnails = [{'url': thumbnail}]
2029
2030 category = microformat.get('category') or search_meta('genre')
2031 channel_id = video_details.get('channelId') \
2032 or microformat.get('externalChannelId') \
2033 or search_meta('channelId')
2034 duration = int_or_none(
2035 video_details.get('lengthSeconds')
2036 or microformat.get('lengthSeconds')) \
2037 or parse_duration(search_meta('duration'))
2038 is_live = video_details.get('isLive')
2039 owner_profile_url = microformat.get('ownerProfileUrl')
2040
2041 info = {
2042 'id': video_id,
2043 'title': self._live_title(video_title) if is_live else video_title,
2044 'formats': formats,
2045 'thumbnails': thumbnails,
2046 'description': video_description,
2047 'upload_date': unified_strdate(
2048 microformat.get('uploadDate')
2049 or search_meta('uploadDate')),
2050 'uploader': video_details['author'],
2051 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
2052 'uploader_url': owner_profile_url,
2053 'channel_id': channel_id,
2054 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None,
2055 'duration': duration,
2056 'view_count': int_or_none(
2057 video_details.get('viewCount')
2058 or microformat.get('viewCount')
2059 or search_meta('interactionCount')),
2060 'average_rating': float_or_none(video_details.get('averageRating')),
2061 'age_limit': 18 if (
2062 microformat.get('isFamilySafe') is False
2063 or search_meta('isFamilyFriendly') == 'false'
2064 or search_meta('og:restrictions:age') == '18+') else 0,
2065 'webpage_url': webpage_url,
2066 'categories': [category] if category else None,
2067 'tags': keywords,
2068 'is_live': is_live,
2069 'playable_in_embed': playability_status.get('playableInEmbed'),
2070 'was_live': video_details.get('isLiveContent'),
2071 }
2072
2073 pctr = try_get(
2074 player_response,
2075 lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
2076 subtitles = {}
2077 if pctr:
2078 def process_language(container, base_url, lang_code, query):
2079 lang_subs = []
2080 for fmt in self._SUBTITLE_FORMATS:
2081 query.update({
2082 'fmt': fmt,
2083 })
2084 lang_subs.append({
2085 'ext': fmt,
2086 'url': update_url_query(base_url, query),
2087 })
2088 container[lang_code] = lang_subs
2089
2090 for caption_track in (pctr.get('captionTracks') or []):
2091 base_url = caption_track.get('baseUrl')
2092 if not base_url:
2093 continue
2094 if caption_track.get('kind') != 'asr':
2095 lang_code = caption_track.get('languageCode')
2096 if not lang_code:
2097 continue
2098 process_language(
2099 subtitles, base_url, lang_code, {})
2100 continue
2101 automatic_captions = {}
2102 for translation_language in (pctr.get('translationLanguages') or []):
2103 translation_language_code = translation_language.get('languageCode')
2104 if not translation_language_code:
2105 continue
2106 process_language(
2107 automatic_captions, base_url, translation_language_code,
2108 {'tlang': translation_language_code})
2109 info['automatic_captions'] = automatic_captions
2110 info['subtitles'] = subtitles
2111
2112 parsed_url = compat_urllib_parse_urlparse(url)
2113 for component in [parsed_url.fragment, parsed_url.query]:
2114 query = compat_parse_qs(component)
2115 for k, v in query.items():
2116 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
2117 d_k += '_time'
2118 if d_k not in info and k in s_ks:
2119 info[d_k] = parse_duration(query[k][0])
2120
2121 # Youtube Music Auto-generated description
2122 if video_description:
2123 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)
2124 if mobj:
2125 release_year = mobj.group('release_year')
2126 release_date = mobj.group('release_date')
2127 if release_date:
2128 release_date = release_date.replace('-', '')
2129 if not release_year:
2130 release_year = release_date[:4]
2131 info.update({
2132 'album': mobj.group('album'.strip()),
2133 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
2134 'track': mobj.group('track').strip(),
2135 'release_date': release_date,
2136 'release_year': int_or_none(release_year),
2137 })
2138
2139 initial_data = None
2140 if webpage:
2141 initial_data = self._extract_yt_initial_variable(
2142 webpage, self._YT_INITIAL_DATA_RE, video_id,
2143 'yt initial data')
2144 if not initial_data:
2145 initial_data = self._call_api(
2146 'next', {'videoId': video_id}, video_id, fatal=False)
2147
2148 if not is_live:
2149 try:
2150 # This will error if there is no livechat
2151 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2152 info['subtitles']['live_chat'] = [{
2153 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
2154 'video_id': video_id,
2155 'ext': 'json',
2156 'protocol': 'youtube_live_chat_replay',
2157 }]
2158 except (KeyError, IndexError, TypeError):
2159 pass
2160
2161 if initial_data:
2162 chapters = self._extract_chapters_from_json(
2163 initial_data, video_id, duration)
2164 if not chapters:
2165 for engagment_pannel in (initial_data.get('engagementPanels') or []):
2166 contents = try_get(
2167 engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'],
2168 list)
2169 if not contents:
2170 continue
2171
2172 def chapter_time(mmlir):
2173 return parse_duration(
2174 get_text(mmlir.get('timeDescription')))
2175
2176 chapters = []
2177 for next_num, content in enumerate(contents, start=1):
2178 mmlir = content.get('macroMarkersListItemRenderer') or {}
2179 start_time = chapter_time(mmlir)
2180 end_time = chapter_time(try_get(
2181 contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \
2182 if next_num < len(contents) else duration
2183 if start_time is None or end_time is None:
2184 continue
2185 chapters.append({
2186 'start_time': start_time,
2187 'end_time': end_time,
2188 'title': get_text(mmlir.get('title')),
2189 })
2190 if chapters:
2191 break
2192 if chapters:
2193 info['chapters'] = chapters
2194
2195 contents = try_get(
2196 initial_data,
2197 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
2198 list) or []
2199 for content in contents:
2200 vpir = content.get('videoPrimaryInfoRenderer')
2201 if vpir:
2202 stl = vpir.get('superTitleLink')
2203 if stl:
2204 stl = get_text(stl)
2205 if try_get(
2206 vpir,
2207 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
2208 info['location'] = stl
2209 else:
2210 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
2211 if mobj:
2212 info.update({
2213 'series': mobj.group(1),
2214 'season_number': int(mobj.group(2)),
2215 'episode_number': int(mobj.group(3)),
2216 })
2217 for tlb in (try_get(
2218 vpir,
2219 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
2220 list) or []):
2221 tbr = tlb.get('toggleButtonRenderer') or {}
2222 for getter, regex in [(
2223 lambda x: x['defaultText']['accessibility']['accessibilityData'],
2224 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
2225 lambda x: x['accessibility'],
2226 lambda x: x['accessibilityData']['accessibilityData'],
2227 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
2228 label = (try_get(tbr, getter, dict) or {}).get('label')
2229 if label:
2230 mobj = re.match(regex, label)
2231 if mobj:
2232 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
2233 break
2234 sbr_tooltip = try_get(
2235 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
2236 if sbr_tooltip:
2237 like_count, dislike_count = sbr_tooltip.split(' / ')
2238 info.update({
2239 'like_count': str_to_int(like_count),
2240 'dislike_count': str_to_int(dislike_count),
2241 })
2242 vsir = content.get('videoSecondaryInfoRenderer')
2243 if vsir:
2244 info['channel'] = get_text(try_get(
2245 vsir,
2246 lambda x: x['owner']['videoOwnerRenderer']['title'],
2247 compat_str))
2248 rows = try_get(
2249 vsir,
2250 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
2251 list) or []
2252 multiple_songs = False
2253 for row in rows:
2254 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2255 multiple_songs = True
2256 break
2257 for row in rows:
2258 mrr = row.get('metadataRowRenderer') or {}
2259 mrr_title = mrr.get('title')
2260 if not mrr_title:
2261 continue
2262 mrr_title = get_text(mrr['title'])
2263 mrr_contents_text = get_text(mrr['contents'][0])
2264 if mrr_title == 'License':
2265 info['license'] = mrr_contents_text
2266 elif not multiple_songs:
2267 if mrr_title == 'Album':
2268 info['album'] = mrr_contents_text
2269 elif mrr_title == 'Artist':
2270 info['artist'] = mrr_contents_text
2271 elif mrr_title == 'Song':
2272 info['track'] = mrr_contents_text
2273
2274 fallbacks = {
2275 'channel': 'uploader',
2276 'channel_id': 'uploader_id',
2277 'channel_url': 'uploader_url',
2278 }
2279 for to, frm in fallbacks.items():
2280 if not info.get(to):
2281 info[to] = info.get(frm)
2282
2283 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
2284 v = info.get(s_k)
2285 if v:
2286 info[d_k] = v
2287
2288 is_private = bool_or_none(video_details.get('isPrivate'))
2289 is_unlisted = bool_or_none(microformat.get('isUnlisted'))
2290 is_membersonly = None
2291 if initial_data and is_private is not None:
2292 is_membersonly = False
2293 contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list)
2294 for content in contents or []:
2295 badges = try_get(content, lambda x: x['videoPrimaryInfoRenderer']['badges'], list)
2296 for badge in badges or []:
2297 label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label']) or ''
2298 if label.lower() == 'members only':
2299 is_membersonly = True
2300 break
2301 if is_membersonly:
2302 break
2303
2304 # TODO: Add this for playlists
2305 info['availability'] = self._availability(
2306 is_private=is_private,
2307 needs_premium=False, # Youtube no longer have premium-only videos?
2308 needs_subscription=is_membersonly,
2309 needs_auth=info['age_limit'] >= 18,
2310 is_unlisted=None if is_private is None else is_unlisted)
2311
2312 # get xsrf for annotations or comments
2313 get_annotations = self._downloader.params.get('writeannotations', False)
2314 get_comments = self._downloader.params.get('getcomments', False)
2315 if get_annotations or get_comments:
2316 xsrf_token = None
2317 ytcfg = self._extract_ytcfg(video_id, webpage)
2318 if ytcfg:
2319 xsrf_token = try_get(ytcfg, lambda x: x['XSRF_TOKEN'], compat_str)
2320 if not xsrf_token:
2321 xsrf_token = self._search_regex(
2322 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>(?:(?!\2).)+)\2',
2323 webpage, 'xsrf token', group='xsrf_token', fatal=False)
2324
2325 # annotations
2326 if get_annotations:
2327 invideo_url = try_get(
2328 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2329 if xsrf_token and invideo_url:
2330 xsrf_field_name = None
2331 if ytcfg:
2332 xsrf_field_name = try_get(ytcfg, lambda x: x['XSRF_FIELD_NAME'], compat_str)
2333 if not xsrf_field_name:
2334 xsrf_field_name = self._search_regex(
2335 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2336 webpage, 'xsrf field name',
2337 group='xsrf_field_name', default='session_token')
2338 info['annotations'] = self._download_webpage(
2339 self._proto_relative_url(invideo_url),
2340 video_id, note='Downloading annotations',
2341 errnote='Unable to download video annotations', fatal=False,
2342 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2343
2344 if get_comments:
2345 info['__post_extractor'] = lambda: self._extract_comments(ytcfg, video_id, contents, webpage, xsrf_token)
2346
2347 self.mark_watched(video_id, player_response)
2348
2349 return info
2350
2351
2352 class YoutubeTabIE(YoutubeBaseInfoExtractor):
2353 IE_DESC = 'YouTube.com tab'
2354 _VALID_URL = r'''(?x)
2355 https?://
2356 (?:\w+\.)?
2357 (?:
2358 youtube(?:kids)?\.com|
2359 invidio\.us
2360 )/
2361 (?:
2362 (?:channel|c|user)/|
2363 (?P<not_channel>
2364 feed/|hashtag/|
2365 (?:playlist|watch)\?.*?\blist=
2366 )|
2367 (?!(?:%s)\b) # Direct URLs
2368 )
2369 (?P<id>[^/?\#&]+)
2370 ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
2371 IE_NAME = 'youtube:tab'
2372
2373 _TESTS = [{
2374 # playlists, multipage
2375 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
2376 'playlist_mincount': 94,
2377 'info_dict': {
2378 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2379 'title': 'Игорь Клейнер - Playlists',
2380 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2381 'uploader': 'Игорь Клейнер',
2382 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2383 },
2384 }, {
2385 # playlists, multipage, different order
2386 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2387 'playlist_mincount': 94,
2388 'info_dict': {
2389 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2390 'title': 'Игорь Клейнер - Playlists',
2391 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2392 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2393 'uploader': 'Игорь Клейнер',
2394 },
2395 }, {
2396 # playlists, singlepage
2397 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2398 'playlist_mincount': 4,
2399 'info_dict': {
2400 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2401 'title': 'ThirstForScience - Playlists',
2402 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
2403 'uploader': 'ThirstForScience',
2404 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2405 }
2406 }, {
2407 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
2408 'only_matching': True,
2409 }, {
2410 # basic, single video playlist
2411 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2412 'info_dict': {
2413 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2414 'uploader': 'Sergey M.',
2415 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2416 'title': 'youtube-dl public playlist',
2417 },
2418 'playlist_count': 1,
2419 }, {
2420 # empty playlist
2421 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2422 'info_dict': {
2423 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2424 'uploader': 'Sergey M.',
2425 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2426 'title': 'youtube-dl empty playlist',
2427 },
2428 'playlist_count': 0,
2429 }, {
2430 # Home tab
2431 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
2432 'info_dict': {
2433 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2434 'title': 'lex will - Home',
2435 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2436 'uploader': 'lex will',
2437 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2438 },
2439 'playlist_mincount': 2,
2440 }, {
2441 # Videos tab
2442 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
2443 'info_dict': {
2444 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2445 'title': 'lex will - Videos',
2446 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2447 'uploader': 'lex will',
2448 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2449 },
2450 'playlist_mincount': 975,
2451 }, {
2452 # Videos tab, sorted by popular
2453 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
2454 'info_dict': {
2455 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2456 'title': 'lex will - Videos',
2457 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2458 'uploader': 'lex will',
2459 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2460 },
2461 'playlist_mincount': 199,
2462 }, {
2463 # Playlists tab
2464 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
2465 'info_dict': {
2466 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2467 'title': 'lex will - Playlists',
2468 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2469 'uploader': 'lex will',
2470 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2471 },
2472 'playlist_mincount': 17,
2473 }, {
2474 # Community tab
2475 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
2476 'info_dict': {
2477 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2478 'title': 'lex will - Community',
2479 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2480 'uploader': 'lex will',
2481 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2482 },
2483 'playlist_mincount': 18,
2484 }, {
2485 # Channels tab
2486 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
2487 'info_dict': {
2488 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2489 'title': 'lex will - Channels',
2490 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2491 'uploader': 'lex will',
2492 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2493 },
2494 'playlist_mincount': 12,
2495 }, {
2496 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2497 'only_matching': True,
2498 }, {
2499 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2500 'only_matching': True,
2501 }, {
2502 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2503 'only_matching': True,
2504 }, {
2505 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2506 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2507 'info_dict': {
2508 'title': '29C3: Not my department',
2509 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2510 'uploader': 'Christiaan008',
2511 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
2512 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
2513 },
2514 'playlist_count': 96,
2515 }, {
2516 'note': 'Large playlist',
2517 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2518 'info_dict': {
2519 'title': 'Uploads from Cauchemar',
2520 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2521 'uploader': 'Cauchemar',
2522 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
2523 },
2524 'playlist_mincount': 1123,
2525 }, {
2526 # even larger playlist, 8832 videos
2527 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
2528 'only_matching': True,
2529 }, {
2530 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2531 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2532 'info_dict': {
2533 'title': 'Uploads from Interstellar Movie',
2534 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2535 'uploader': 'Interstellar Movie',
2536 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
2537 },
2538 'playlist_mincount': 21,
2539 }, {
2540 # https://github.com/ytdl-org/youtube-dl/issues/21844
2541 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2542 'info_dict': {
2543 'title': 'Data Analysis with Dr Mike Pound',
2544 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2545 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
2546 'uploader': 'Computerphile',
2547 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
2548 },
2549 'playlist_mincount': 11,
2550 }, {
2551 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2552 'only_matching': True,
2553 }, {
2554 # Playlist URL that does not actually serve a playlist
2555 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2556 'info_dict': {
2557 'id': 'FqZTN594JQw',
2558 'ext': 'webm',
2559 'title': "Smiley's People 01 detective, Adventure Series, Action",
2560 'uploader': 'STREEM',
2561 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2562 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2563 'upload_date': '20150526',
2564 'license': 'Standard YouTube License',
2565 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2566 'categories': ['People & Blogs'],
2567 'tags': list,
2568 'view_count': int,
2569 'like_count': int,
2570 'dislike_count': int,
2571 },
2572 'params': {
2573 'skip_download': True,
2574 },
2575 'skip': 'This video is not available.',
2576 'add_ie': [YoutubeIE.ie_key()],
2577 }, {
2578 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2579 'only_matching': True,
2580 }, {
2581 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
2582 'only_matching': True,
2583 }, {
2584 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
2585 'info_dict': {
2586 'id': '9Auq9mYxFEE',
2587 'ext': 'mp4',
2588 'title': compat_str,
2589 'uploader': 'Sky News',
2590 'uploader_id': 'skynews',
2591 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
2592 'upload_date': '20191102',
2593 'description': 'md5:85ddd75d888674631aaf9599a9a0b0ae',
2594 'categories': ['News & Politics'],
2595 'tags': list,
2596 'like_count': int,
2597 'dislike_count': int,
2598 },
2599 'params': {
2600 'skip_download': True,
2601 },
2602 }, {
2603 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2604 'info_dict': {
2605 'id': 'a48o2S1cPoo',
2606 'ext': 'mp4',
2607 'title': 'The Young Turks - Live Main Show',
2608 'uploader': 'The Young Turks',
2609 'uploader_id': 'TheYoungTurks',
2610 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2611 'upload_date': '20150715',
2612 'license': 'Standard YouTube License',
2613 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2614 'categories': ['News & Politics'],
2615 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2616 'like_count': int,
2617 'dislike_count': int,
2618 },
2619 'params': {
2620 'skip_download': True,
2621 },
2622 'only_matching': True,
2623 }, {
2624 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2625 'only_matching': True,
2626 }, {
2627 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2628 'only_matching': True,
2629 }, {
2630 'url': 'https://www.youtube.com/feed/trending',
2631 'only_matching': True,
2632 }, {
2633 # needs auth
2634 'url': 'https://www.youtube.com/feed/library',
2635 'only_matching': True,
2636 }, {
2637 # needs auth
2638 'url': 'https://www.youtube.com/feed/history',
2639 'only_matching': True,
2640 }, {
2641 # needs auth
2642 'url': 'https://www.youtube.com/feed/subscriptions',
2643 'only_matching': True,
2644 }, {
2645 # needs auth
2646 'url': 'https://www.youtube.com/feed/watch_later',
2647 'only_matching': True,
2648 }, {
2649 # no longer available?
2650 'url': 'https://www.youtube.com/feed/recommended',
2651 'only_matching': True,
2652 }, {
2653 # inline playlist with not always working continuations
2654 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
2655 'only_matching': True,
2656 }, {
2657 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8',
2658 'only_matching': True,
2659 }, {
2660 'url': 'https://www.youtube.com/course',
2661 'only_matching': True,
2662 }, {
2663 'url': 'https://www.youtube.com/zsecurity',
2664 'only_matching': True,
2665 }, {
2666 'url': 'http://www.youtube.com/NASAgovVideo/videos',
2667 'only_matching': True,
2668 }, {
2669 'url': 'https://www.youtube.com/TheYoungTurks/live',
2670 'only_matching': True,
2671 }]
2672
2673 @classmethod
2674 def suitable(cls, url):
2675 return False if YoutubeIE.suitable(url) else super(
2676 YoutubeTabIE, cls).suitable(url)
2677
2678 def _extract_channel_id(self, webpage):
2679 channel_id = self._html_search_meta(
2680 'channelId', webpage, 'channel id', default=None)
2681 if channel_id:
2682 return channel_id
2683 channel_url = self._html_search_meta(
2684 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
2685 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
2686 'twitter:app:url:googleplay'), webpage, 'channel url')
2687 return self._search_regex(
2688 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
2689 channel_url, 'channel id')
2690
2691 @staticmethod
2692 def _extract_basic_item_renderer(item):
2693 # Modified from _extract_grid_item_renderer
2694 known_renderers = (
2695 'playlistRenderer', 'videoRenderer', 'channelRenderer',
2696 'gridPlaylistRenderer', 'gridVideoRenderer', 'gridChannelRenderer'
2697 )
2698 for key, renderer in item.items():
2699 if key not in known_renderers:
2700 continue
2701 return renderer
2702
2703 def _grid_entries(self, grid_renderer):
2704 for item in grid_renderer['items']:
2705 if not isinstance(item, dict):
2706 continue
2707 renderer = self._extract_basic_item_renderer(item)
2708 if not isinstance(renderer, dict):
2709 continue
2710 title = try_get(
2711 renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2712 # playlist
2713 playlist_id = renderer.get('playlistId')
2714 if playlist_id:
2715 yield self.url_result(
2716 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2717 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
2718 video_title=title)
2719 # video
2720 video_id = renderer.get('videoId')
2721 if video_id:
2722 yield self._extract_video(renderer)
2723 # channel
2724 channel_id = renderer.get('channelId')
2725 if channel_id:
2726 title = try_get(
2727 renderer, lambda x: x['title']['simpleText'], compat_str)
2728 yield self.url_result(
2729 'https://www.youtube.com/channel/%s' % channel_id,
2730 ie=YoutubeTabIE.ie_key(), video_title=title)
2731
2732 def _shelf_entries_from_content(self, shelf_renderer):
2733 content = shelf_renderer.get('content')
2734 if not isinstance(content, dict):
2735 return
2736 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
2737 if renderer:
2738 # TODO: add support for nested playlists so each shelf is processed
2739 # as separate playlist
2740 # TODO: this includes only first N items
2741 for entry in self._grid_entries(renderer):
2742 yield entry
2743 renderer = content.get('horizontalListRenderer')
2744 if renderer:
2745 # TODO
2746 pass
2747
2748 def _shelf_entries(self, shelf_renderer, skip_channels=False):
2749 ep = try_get(
2750 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
2751 compat_str)
2752 shelf_url = urljoin('https://www.youtube.com', ep)
2753 if shelf_url:
2754 # Skipping links to another channels, note that checking for
2755 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
2756 # will not work
2757 if skip_channels and '/channels?' in shelf_url:
2758 return
2759 title = try_get(
2760 shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2761 yield self.url_result(shelf_url, video_title=title)
2762 # Shelf may not contain shelf URL, fallback to extraction from content
2763 for entry in self._shelf_entries_from_content(shelf_renderer):
2764 yield entry
2765
2766 def _playlist_entries(self, video_list_renderer):
2767 for content in video_list_renderer['contents']:
2768 if not isinstance(content, dict):
2769 continue
2770 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
2771 if not isinstance(renderer, dict):
2772 continue
2773 video_id = renderer.get('videoId')
2774 if not video_id:
2775 continue
2776 yield self._extract_video(renderer)
2777
2778 def _rich_entries(self, rich_grid_renderer):
2779 renderer = try_get(
2780 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
2781 video_id = renderer.get('videoId')
2782 if not video_id:
2783 return
2784 yield self._extract_video(renderer)
2785
2786 def _video_entry(self, video_renderer):
2787 video_id = video_renderer.get('videoId')
2788 if video_id:
2789 return self._extract_video(video_renderer)
2790
2791 def _post_thread_entries(self, post_thread_renderer):
2792 post_renderer = try_get(
2793 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
2794 if not post_renderer:
2795 return
2796 # video attachment
2797 video_renderer = try_get(
2798 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict)
2799 video_id = None
2800 if video_renderer:
2801 entry = self._video_entry(video_renderer)
2802 if entry:
2803 yield entry
2804 # inline video links
2805 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
2806 for run in runs:
2807 if not isinstance(run, dict):
2808 continue
2809 ep_url = try_get(
2810 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
2811 if not ep_url:
2812 continue
2813 if not YoutubeIE.suitable(ep_url):
2814 continue
2815 ep_video_id = YoutubeIE._match_id(ep_url)
2816 if video_id == ep_video_id:
2817 continue
2818 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=video_id)
2819
2820 def _post_thread_continuation_entries(self, post_thread_continuation):
2821 contents = post_thread_continuation.get('contents')
2822 if not isinstance(contents, list):
2823 return
2824 for content in contents:
2825 renderer = content.get('backstagePostThreadRenderer')
2826 if not isinstance(renderer, dict):
2827 continue
2828 for entry in self._post_thread_entries(renderer):
2829 yield entry
2830
2831 @staticmethod
2832 def _build_continuation_query(continuation, ctp=None):
2833 query = {
2834 'ctoken': continuation,
2835 'continuation': continuation,
2836 }
2837 if ctp:
2838 query['itct'] = ctp
2839 return query
2840
2841 @staticmethod
2842 def _extract_next_continuation_data(renderer):
2843 next_continuation = try_get(
2844 renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict)
2845 if not next_continuation:
2846 return
2847 continuation = next_continuation.get('continuation')
2848 if not continuation:
2849 return
2850 ctp = next_continuation.get('clickTrackingParams')
2851 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2852
2853 @classmethod
2854 def _extract_continuation(cls, renderer):
2855 next_continuation = cls._extract_next_continuation_data(renderer)
2856 if next_continuation:
2857 return next_continuation
2858 contents = []
2859 for key in ('contents', 'items'):
2860 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
2861 for content in contents:
2862 if not isinstance(content, dict):
2863 continue
2864 continuation_ep = try_get(
2865 content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
2866 dict)
2867 if not continuation_ep:
2868 continue
2869 continuation = try_get(
2870 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
2871 if not continuation:
2872 continue
2873 ctp = continuation_ep.get('clickTrackingParams')
2874 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2875
2876 def _entries(self, tab, item_id, identity_token, account_syncid):
2877
2878 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
2879 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
2880 for content in contents:
2881 if not isinstance(content, dict):
2882 continue
2883 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
2884 if not is_renderer:
2885 renderer = content.get('richItemRenderer')
2886 if renderer:
2887 for entry in self._rich_entries(renderer):
2888 yield entry
2889 continuation_list[0] = self._extract_continuation(parent_renderer)
2890 continue
2891 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
2892 for isr_content in isr_contents:
2893 if not isinstance(isr_content, dict):
2894 continue
2895
2896 known_renderers = {
2897 'playlistVideoListRenderer': self._playlist_entries,
2898 'gridRenderer': self._grid_entries,
2899 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'),
2900 'backstagePostThreadRenderer': self._post_thread_entries,
2901 'videoRenderer': lambda x: [self._video_entry(x)],
2902 }
2903 for key, renderer in isr_content.items():
2904 if key not in known_renderers:
2905 continue
2906 for entry in known_renderers[key](renderer):
2907 if entry:
2908 yield entry
2909 continuation_list[0] = self._extract_continuation(renderer)
2910 break
2911
2912 if not continuation_list[0]:
2913 continuation_list[0] = self._extract_continuation(is_renderer)
2914
2915 if not continuation_list[0]:
2916 continuation_list[0] = self._extract_continuation(parent_renderer)
2917
2918 continuation_list = [None] # Python 2 doesnot support nonlocal
2919 tab_content = try_get(tab, lambda x: x['content'], dict)
2920 if not tab_content:
2921 return
2922 parent_renderer = (
2923 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
2924 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
2925 for entry in extract_entries(parent_renderer):
2926 yield entry
2927 continuation = continuation_list[0]
2928
2929 headers = {
2930 'x-youtube-client-name': '1',
2931 'x-youtube-client-version': '2.20201112.04.01',
2932 }
2933 if identity_token:
2934 headers['x-youtube-identity-token'] = identity_token
2935
2936 if account_syncid:
2937 headers['X-Goog-PageId'] = account_syncid
2938 headers['X-Goog-AuthUser'] = 0
2939
2940 for page_num in itertools.count(1):
2941 if not continuation:
2942 break
2943 retries = self._downloader.params.get('extractor_retries', 3)
2944 count = -1
2945 last_error = None
2946 while count < retries:
2947 count += 1
2948 if last_error:
2949 self.report_warning('%s. Retrying ...' % last_error)
2950 try:
2951 response = self._call_api(
2952 ep="browse", fatal=True, headers=headers,
2953 video_id='%s page %s' % (item_id, page_num),
2954 query={
2955 'continuation': continuation['continuation'],
2956 'clickTracking': {'clickTrackingParams': continuation['itct']},
2957 },
2958 note='Downloading API JSON%s' % (' (retry #%d)' % count if count else ''))
2959 except ExtractorError as e:
2960 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404):
2961 # Downloading page may result in intermittent 5xx HTTP error
2962 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
2963 last_error = 'HTTP Error %s' % e.cause.code
2964 if count < retries:
2965 continue
2966 raise
2967 else:
2968 # Youtube sometimes sends incomplete data
2969 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
2970 if dict_get(response,
2971 ('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints')):
2972 break
2973
2974 # Youtube may send alerts if there was an issue with the continuation page
2975 self._extract_alerts(response, expected=False)
2976
2977 last_error = 'Incomplete data received'
2978 if count >= retries:
2979 self._downloader.report_error(last_error)
2980
2981 if not response:
2982 break
2983
2984 known_continuation_renderers = {
2985 'playlistVideoListContinuation': self._playlist_entries,
2986 'gridContinuation': self._grid_entries,
2987 'itemSectionContinuation': self._post_thread_continuation_entries,
2988 'sectionListContinuation': extract_entries, # for feeds
2989 }
2990 continuation_contents = try_get(
2991 response, lambda x: x['continuationContents'], dict) or {}
2992 continuation_renderer = None
2993 for key, value in continuation_contents.items():
2994 if key not in known_continuation_renderers:
2995 continue
2996 continuation_renderer = value
2997 continuation_list = [None]
2998 for entry in known_continuation_renderers[key](continuation_renderer):
2999 yield entry
3000 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
3001 break
3002 if continuation_renderer:
3003 continue
3004
3005 known_renderers = {
3006 'gridPlaylistRenderer': (self._grid_entries, 'items'),
3007 'gridVideoRenderer': (self._grid_entries, 'items'),
3008 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
3009 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
3010 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
3011 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
3012 }
3013 continuation_items = try_get(
3014 response,
3015 lambda x: dict_get(x, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))[0]['appendContinuationItemsAction']['continuationItems'], list)
3016 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
3017 video_items_renderer = None
3018 for key, value in continuation_item.items():
3019 if key not in known_renderers:
3020 continue
3021 video_items_renderer = {known_renderers[key][1]: continuation_items}
3022 continuation_list = [None]
3023 for entry in known_renderers[key][0](video_items_renderer):
3024 yield entry
3025 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3026 break
3027 if video_items_renderer:
3028 continue
3029 break
3030
3031 @staticmethod
3032 def _extract_selected_tab(tabs):
3033 for tab in tabs:
3034 if try_get(tab, lambda x: x['tabRenderer']['selected'], bool):
3035 return tab['tabRenderer']
3036 else:
3037 raise ExtractorError('Unable to find selected tab')
3038
3039 @staticmethod
3040 def _extract_uploader(data):
3041 uploader = {}
3042 sidebar_renderer = try_get(
3043 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3044 if sidebar_renderer:
3045 for item in sidebar_renderer:
3046 if not isinstance(item, dict):
3047 continue
3048 renderer = item.get('playlistSidebarSecondaryInfoRenderer')
3049 if not isinstance(renderer, dict):
3050 continue
3051 owner = try_get(
3052 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3053 if owner:
3054 uploader['uploader'] = owner.get('text')
3055 uploader['uploader_id'] = try_get(
3056 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3057 uploader['uploader_url'] = urljoin(
3058 'https://www.youtube.com/',
3059 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3060 return {k: v for k, v in uploader.items() if v is not None}
3061
3062 def _extract_from_tabs(self, item_id, webpage, data, tabs):
3063 playlist_id = title = description = channel_url = channel_name = channel_id = None
3064 thumbnails_list = tags = []
3065
3066 selected_tab = self._extract_selected_tab(tabs)
3067 renderer = try_get(
3068 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3069 if renderer:
3070 channel_name = renderer.get('title')
3071 channel_url = renderer.get('channelUrl')
3072 channel_id = renderer.get('externalId')
3073
3074 if not renderer:
3075 renderer = try_get(
3076 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3077 if renderer:
3078 title = renderer.get('title')
3079 description = renderer.get('description', '')
3080 playlist_id = channel_id
3081 tags = renderer.get('keywords', '').split()
3082 thumbnails_list = (
3083 try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
3084 or try_get(
3085 data,
3086 lambda x: x['sidebar']['playlistSidebarRenderer']['items'][0]['playlistSidebarPrimaryInfoRenderer']['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
3087 list)
3088 or [])
3089
3090 thumbnails = []
3091 for t in thumbnails_list:
3092 if not isinstance(t, dict):
3093 continue
3094 thumbnail_url = url_or_none(t.get('url'))
3095 if not thumbnail_url:
3096 continue
3097 thumbnails.append({
3098 'url': thumbnail_url,
3099 'width': int_or_none(t.get('width')),
3100 'height': int_or_none(t.get('height')),
3101 })
3102
3103 if playlist_id is None:
3104 playlist_id = item_id
3105 if title is None:
3106 title = playlist_id
3107 title += format_field(selected_tab, 'title', ' - %s')
3108
3109 metadata = {
3110 'playlist_id': playlist_id,
3111 'playlist_title': title,
3112 'playlist_description': description,
3113 'uploader': channel_name,
3114 'uploader_id': channel_id,
3115 'uploader_url': channel_url,
3116 'thumbnails': thumbnails,
3117 'tags': tags,
3118 }
3119 if not channel_id:
3120 metadata.update(self._extract_uploader(data))
3121 metadata.update({
3122 'channel': metadata['uploader'],
3123 'channel_id': metadata['uploader_id'],
3124 'channel_url': metadata['uploader_url']})
3125 return self.playlist_result(
3126 self._entries(
3127 selected_tab, playlist_id,
3128 self._extract_identity_token(webpage, item_id),
3129 self._extract_account_syncid(data)),
3130 **metadata)
3131
3132 def _extract_mix_playlist(self, playlist, playlist_id):
3133 first_id = last_id = None
3134 for page_num in itertools.count(1):
3135 videos = list(self._playlist_entries(playlist))
3136 if not videos:
3137 return
3138 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
3139 if start >= len(videos):
3140 return
3141 for video in videos[start:]:
3142 if video['id'] == first_id:
3143 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
3144 return
3145 yield video
3146 first_id = first_id or videos[0]['id']
3147 last_id = videos[-1]['id']
3148
3149 _, data = self._extract_webpage(
3150 'https://www.youtube.com/watch?list=%s&v=%s' % (playlist_id, last_id),
3151 '%s page %d' % (playlist_id, page_num))
3152 playlist = try_get(
3153 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3154
3155 def _extract_from_playlist(self, item_id, url, data, playlist):
3156 title = playlist.get('title') or try_get(
3157 data, lambda x: x['titleText']['simpleText'], compat_str)
3158 playlist_id = playlist.get('playlistId') or item_id
3159
3160 # Delegating everything except mix playlists to regular tab-based playlist URL
3161 playlist_url = urljoin(url, try_get(
3162 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3163 compat_str))
3164 if playlist_url and playlist_url != url:
3165 return self.url_result(
3166 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3167 video_title=title)
3168
3169 return self.playlist_result(
3170 self._extract_mix_playlist(playlist, playlist_id),
3171 playlist_id=playlist_id, playlist_title=title)
3172
3173 def _extract_alerts(self, data, expected=False):
3174
3175 def _real_extract_alerts():
3176 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
3177 if not isinstance(alert_dict, dict):
3178 continue
3179 for alert in alert_dict.values():
3180 alert_type = alert.get('type')
3181 if not alert_type:
3182 continue
3183 message = try_get(alert, lambda x: x['text']['simpleText'], compat_str)
3184 if message:
3185 yield alert_type, message
3186 for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
3187 message = try_get(run, lambda x: x['text'], compat_str)
3188 if message:
3189 yield alert_type, message
3190
3191 err_msg = None
3192 for alert_type, alert_message in _real_extract_alerts():
3193 if alert_type.lower() == 'error':
3194 if err_msg:
3195 self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
3196 err_msg = alert_message
3197 else:
3198 self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
3199
3200 if err_msg:
3201 raise ExtractorError('YouTube said: %s' % err_msg, expected=expected)
3202
3203 def _extract_webpage(self, url, item_id):
3204 retries = self._downloader.params.get('extractor_retries', 3)
3205 count = -1
3206 last_error = 'Incomplete yt initial data recieved'
3207 while count < retries:
3208 count += 1
3209 # Sometimes youtube returns a webpage with incomplete ytInitialData
3210 # See: https://github.com/yt-dlp/yt-dlp/issues/116
3211 if count:
3212 self.report_warning('%s. Retrying ...' % last_error)
3213 webpage = self._download_webpage(
3214 url, item_id,
3215 'Downloading webpage%s' % (' (retry #%d)' % count if count else ''))
3216 data = self._extract_yt_initial_data(item_id, webpage)
3217 self._extract_alerts(data, expected=True)
3218 if data.get('contents') or data.get('currentVideoEndpoint'):
3219 break
3220 if count >= retries:
3221 self._downloader.report_error(last_error)
3222 return webpage, data
3223
3224 def _real_extract(self, url):
3225 item_id = self._match_id(url)
3226 url = compat_urlparse.urlunparse(
3227 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
3228
3229 # This is not matched in a channel page with a tab selected
3230 mobj = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
3231 mobj = mobj.groupdict() if mobj else {}
3232 if mobj and not mobj.get('not_channel'):
3233 self._downloader.report_warning(
3234 'A channel/user page was given. All the channel\'s videos will be downloaded. '
3235 'To download only the videos in the home page, add a "/featured" to the URL')
3236 url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '')
3237
3238 # Handle both video/playlist URLs
3239 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3240 video_id = qs.get('v', [None])[0]
3241 playlist_id = qs.get('list', [None])[0]
3242
3243 if not video_id and (mobj.get('not_channel') or '').startswith('watch'):
3244 if not playlist_id:
3245 # If there is neither video or playlist ids,
3246 # youtube redirects to home page, which is undesirable
3247 raise ExtractorError('Unable to recognize tab page')
3248 self._downloader.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
3249 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
3250
3251 if video_id and playlist_id:
3252 if self._downloader.params.get('noplaylist'):
3253 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3254 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3255 self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
3256
3257 webpage, data = self._extract_webpage(url, item_id)
3258
3259 tabs = try_get(
3260 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
3261 if tabs:
3262 return self._extract_from_tabs(item_id, webpage, data, tabs)
3263
3264 playlist = try_get(
3265 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3266 if playlist:
3267 return self._extract_from_playlist(item_id, url, data, playlist)
3268
3269 video_id = try_get(
3270 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
3271 compat_str) or video_id
3272 if video_id:
3273 self._downloader.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
3274 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3275
3276 raise ExtractorError('Unable to recognize tab page')
3277
3278
3279 class YoutubePlaylistIE(InfoExtractor):
3280 IE_DESC = 'YouTube.com playlists'
3281 _VALID_URL = r'''(?x)(?:
3282 (?:https?://)?
3283 (?:\w+\.)?
3284 (?:
3285 (?:
3286 youtube(?:kids)?\.com|
3287 invidio\.us
3288 )
3289 /.*?\?.*?\blist=
3290 )?
3291 (?P<id>%(playlist_id)s)
3292 )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3293 IE_NAME = 'youtube:playlist'
3294 _TESTS = [{
3295 'note': 'issue #673',
3296 'url': 'PLBB231211A4F62143',
3297 'info_dict': {
3298 'title': '[OLD]Team Fortress 2 (Class-based LP)',
3299 'id': 'PLBB231211A4F62143',
3300 'uploader': 'Wickydoo',
3301 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
3302 },
3303 'playlist_mincount': 29,
3304 }, {
3305 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3306 'info_dict': {
3307 'title': 'YDL_safe_search',
3308 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3309 },
3310 'playlist_count': 2,
3311 'skip': 'This playlist is private',
3312 }, {
3313 'note': 'embedded',
3314 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3315 'playlist_count': 4,
3316 'info_dict': {
3317 'title': 'JODA15',
3318 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3319 'uploader': 'milan',
3320 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
3321 }
3322 }, {
3323 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3324 'playlist_mincount': 982,
3325 'info_dict': {
3326 'title': '2018 Chinese New Singles (11/6 updated)',
3327 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3328 'uploader': 'LBK',
3329 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
3330 }
3331 }, {
3332 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
3333 'only_matching': True,
3334 }, {
3335 # music album playlist
3336 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
3337 'only_matching': True,
3338 }]
3339
3340 @classmethod
3341 def suitable(cls, url):
3342 return False if YoutubeTabIE.suitable(url) else super(
3343 YoutubePlaylistIE, cls).suitable(url)
3344
3345 def _real_extract(self, url):
3346 playlist_id = self._match_id(url)
3347 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3348 if not qs:
3349 qs = {'list': playlist_id}
3350 return self.url_result(
3351 update_url_query('https://www.youtube.com/playlist', qs),
3352 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3353
3354
3355 class YoutubeYtBeIE(InfoExtractor):
3356 IE_DESC = 'youtu.be'
3357 _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}
3358 _TESTS = [{
3359 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
3360 'info_dict': {
3361 'id': 'yeWKywCrFtk',
3362 'ext': 'mp4',
3363 'title': 'Small Scale Baler and Braiding Rugs',
3364 'uploader': 'Backus-Page House Museum',
3365 'uploader_id': 'backuspagemuseum',
3366 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
3367 'upload_date': '20161008',
3368 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
3369 'categories': ['Nonprofits & Activism'],
3370 'tags': list,
3371 'like_count': int,
3372 'dislike_count': int,
3373 },
3374 'params': {
3375 'noplaylist': True,
3376 'skip_download': True,
3377 },
3378 }, {
3379 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
3380 'only_matching': True,
3381 }]
3382
3383 def _real_extract(self, url):
3384 mobj = re.match(self._VALID_URL, url)
3385 video_id = mobj.group('id')
3386 playlist_id = mobj.group('playlist_id')
3387 return self.url_result(
3388 update_url_query('https://www.youtube.com/watch', {
3389 'v': video_id,
3390 'list': playlist_id,
3391 'feature': 'youtu.be',
3392 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3393
3394
3395 class YoutubeYtUserIE(InfoExtractor):
3396 IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword'
3397 _VALID_URL = r'ytuser:(?P<id>.+)'
3398 _TESTS = [{
3399 'url': 'ytuser:phihag',
3400 'only_matching': True,
3401 }]
3402
3403 def _real_extract(self, url):
3404 user_id = self._match_id(url)
3405 return self.url_result(
3406 'https://www.youtube.com/user/%s' % user_id,
3407 ie=YoutubeTabIE.ie_key(), video_id=user_id)
3408
3409
3410 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3411 IE_NAME = 'youtube:favorites'
3412 IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)'
3413 _VALID_URL = r':ytfav(?:ou?rite)?s?'
3414 _LOGIN_REQUIRED = True
3415 _TESTS = [{
3416 'url': ':ytfav',
3417 'only_matching': True,
3418 }, {
3419 'url': ':ytfavorites',
3420 'only_matching': True,
3421 }]
3422
3423 def _real_extract(self, url):
3424 return self.url_result(
3425 'https://www.youtube.com/playlist?list=LL',
3426 ie=YoutubeTabIE.ie_key())
3427
3428
3429 class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
3430 IE_DESC = 'YouTube.com searches, "ytsearch" keyword'
3431 # there doesn't appear to be a real limit, for example if you search for
3432 # 'python' you get more than 8.000.000 results
3433 _MAX_RESULTS = float('inf')
3434 IE_NAME = 'youtube:search'
3435 _SEARCH_KEY = 'ytsearch'
3436 _SEARCH_PARAMS = None
3437 _TESTS = []
3438
3439 def _entries(self, query, n):
3440 data = {'query': query}
3441 if self._SEARCH_PARAMS:
3442 data['params'] = self._SEARCH_PARAMS
3443 total = 0
3444 for page_num in itertools.count(1):
3445 search = self._call_api(
3446 ep='search', video_id='query "%s"' % query, fatal=False,
3447 note='Downloading page %s' % page_num, query=data)
3448 if not search:
3449 break
3450 slr_contents = try_get(
3451 search,
3452 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
3453 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
3454 list)
3455 if not slr_contents:
3456 break
3457
3458 # Youtube sometimes adds promoted content to searches,
3459 # changing the index location of videos and token.
3460 # So we search through all entries till we find them.
3461 continuation_token = None
3462 for slr_content in slr_contents:
3463 if continuation_token is None:
3464 continuation_token = try_get(
3465 slr_content,
3466 lambda x: x['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'],
3467 compat_str)
3468
3469 isr_contents = try_get(
3470 slr_content,
3471 lambda x: x['itemSectionRenderer']['contents'],
3472 list)
3473 if not isr_contents:
3474 continue
3475 for content in isr_contents:
3476 if not isinstance(content, dict):
3477 continue
3478 video = content.get('videoRenderer')
3479 if not isinstance(video, dict):
3480 continue
3481 video_id = video.get('videoId')
3482 if not video_id:
3483 continue
3484
3485 yield self._extract_video(video)
3486 total += 1
3487 if total == n:
3488 return
3489
3490 if not continuation_token:
3491 break
3492 data['continuation'] = continuation_token
3493
3494 def _get_n_results(self, query, n):
3495 """Get a specified number of results for a query"""
3496 return self.playlist_result(self._entries(query, n), query)
3497
3498
3499 class YoutubeSearchDateIE(YoutubeSearchIE):
3500 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3501 _SEARCH_KEY = 'ytsearchdate'
3502 IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword'
3503 _SEARCH_PARAMS = 'CAI%3D'
3504
3505
3506 class YoutubeSearchURLIE(YoutubeSearchIE):
3507 IE_DESC = 'YouTube.com search URLs'
3508 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
3509 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
3510 # _MAX_RESULTS = 100
3511 _TESTS = [{
3512 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3513 'playlist_mincount': 5,
3514 'info_dict': {
3515 'title': 'youtube-dl test video',
3516 }
3517 }, {
3518 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3519 'only_matching': True,
3520 }]
3521
3522 @classmethod
3523 def _make_valid_url(cls):
3524 return cls._VALID_URL
3525
3526 def _real_extract(self, url):
3527 qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
3528 query = (qs.get('search_query') or qs.get('q'))[0]
3529 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
3530 return self._get_n_results(query, self._MAX_RESULTS)
3531
3532
3533 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
3534 """
3535 Base class for feed extractors
3536 Subclasses must define the _FEED_NAME property.
3537 """
3538 _LOGIN_REQUIRED = True
3539 _TESTS = []
3540
3541 @property
3542 def IE_NAME(self):
3543 return 'youtube:%s' % self._FEED_NAME
3544
3545 def _real_initialize(self):
3546 self._login()
3547
3548 def _real_extract(self, url):
3549 return self.url_result(
3550 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3551 ie=YoutubeTabIE.ie_key())
3552
3553
3554 class YoutubeWatchLaterIE(InfoExtractor):
3555 IE_NAME = 'youtube:watchlater'
3556 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3557 _VALID_URL = r':ytwatchlater'
3558 _TESTS = [{
3559 'url': ':ytwatchlater',
3560 'only_matching': True,
3561 }]
3562
3563 def _real_extract(self, url):
3564 return self.url_result(
3565 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
3566
3567
3568 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3569 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3570 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
3571 _FEED_NAME = 'recommended'
3572 _TESTS = [{
3573 'url': ':ytrec',
3574 'only_matching': True,
3575 }, {
3576 'url': ':ytrecommended',
3577 'only_matching': True,
3578 }, {
3579 'url': 'https://youtube.com',
3580 'only_matching': True,
3581 }]
3582
3583
3584 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3585 IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)'
3586 _VALID_URL = r':ytsub(?:scription)?s?'
3587 _FEED_NAME = 'subscriptions'
3588 _TESTS = [{
3589 'url': ':ytsubs',
3590 'only_matching': True,
3591 }, {
3592 'url': ':ytsubscriptions',
3593 'only_matching': True,
3594 }]
3595
3596
3597 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3598 IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)'
3599 _VALID_URL = r':ythis(?:tory)?'
3600 _FEED_NAME = 'history'
3601 _TESTS = [{
3602 'url': ':ythistory',
3603 'only_matching': True,
3604 }]
3605
3606
3607 class YoutubeTruncatedURLIE(InfoExtractor):
3608 IE_NAME = 'youtube:truncated_url'
3609 IE_DESC = False # Do not list
3610 _VALID_URL = r'''(?x)
3611 (?:https?://)?
3612 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3613 (?:watch\?(?:
3614 feature=[a-z_]+|
3615 annotation_id=annotation_[^&]+|
3616 x-yt-cl=[0-9]+|
3617 hl=[^&]*|
3618 t=[0-9]+
3619 )?
3620 |
3621 attribution_link\?a=[^&]+
3622 )
3623 $
3624 '''
3625
3626 _TESTS = [{
3627 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3628 'only_matching': True,
3629 }, {
3630 'url': 'https://www.youtube.com/watch?',
3631 'only_matching': True,
3632 }, {
3633 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3634 'only_matching': True,
3635 }, {
3636 'url': 'https://www.youtube.com/watch?feature=foo',
3637 'only_matching': True,
3638 }, {
3639 'url': 'https://www.youtube.com/watch?hl=en-GB',
3640 'only_matching': True,
3641 }, {
3642 'url': 'https://www.youtube.com/watch?t=2372',
3643 'only_matching': True,
3644 }]
3645
3646 def _real_extract(self, url):
3647 raise ExtractorError(
3648 'Did you forget to quote the URL? Remember that & is a meta '
3649 'character in most shells, so you want to put the URL in quotes, '
3650 'like youtube-dl '
3651 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3652 ' or simply youtube-dl BaW_jenozKc .',
3653 expected=True)
3654
3655
3656 class YoutubeTruncatedIDIE(InfoExtractor):
3657 IE_NAME = 'youtube:truncated_id'
3658 IE_DESC = False # Do not list
3659 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3660
3661 _TESTS = [{
3662 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3663 'only_matching': True,
3664 }]
3665
3666 def _real_extract(self, url):
3667 video_id = self._match_id(url)
3668 raise ExtractorError(
3669 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3670 expected=True)
fork of https://github.com/yt-dlp/yt-dlp/