]> jfr.im git - yt-dlp.git/blob - youtube_dlc/extractor/youtube.py
Readme changes
[yt-dlp.git] / youtube_dlc / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
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 ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_kwargs,
20 compat_parse_qs,
21 compat_urllib_parse_unquote,
22 compat_urllib_parse_unquote_plus,
23 compat_urllib_parse_urlencode,
24 compat_urllib_parse_urlparse,
25 compat_urlparse,
26 compat_str,
27 )
28 from ..utils import (
29 bool_or_none,
30 clean_html,
31 error_to_compat_str,
32 ExtractorError,
33 float_or_none,
34 get_element_by_id,
35 int_or_none,
36 mimetype2ext,
37 parse_codecs,
38 parse_count,
39 parse_duration,
40 remove_quotes,
41 remove_start,
42 smuggle_url,
43 str_or_none,
44 str_to_int,
45 try_get,
46 unescapeHTML,
47 unified_strdate,
48 unsmuggle_url,
49 update_url_query,
50 uppercase_escape,
51 url_or_none,
52 urlencode_postdata,
53 urljoin,
54 )
55
56
57 class YoutubeBaseInfoExtractor(InfoExtractor):
58 """Provide base functions for Youtube extractors"""
59 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
60 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
61
62 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
63 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
64 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
65
66 _RESERVED_NAMES = (
67 r'course|embed|channel|c|user|playlist|watch|w|results|storefront|oops|'
68 r'shared|index|account|reporthistory|t/terms|about|upload|signin|logout|'
69 r'feed/(watch_later|history|subscriptions|library|trending|recommended)')
70
71 _NETRC_MACHINE = 'youtube'
72 # If True it will raise an error if no login info is provided
73 _LOGIN_REQUIRED = False
74
75 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
76
77 _YOUTUBE_CLIENT_HEADERS = {
78 'x-youtube-client-name': '1',
79 'x-youtube-client-version': '1.20200609.04.02',
80 }
81
82 def _set_language(self):
83 self._set_cookie(
84 '.youtube.com', 'PREF', 'f1=50000000&f6=8&hl=en',
85 # YouTube sets the expire time to about two months
86 expire_time=time.time() + 2 * 30 * 24 * 3600)
87
88 def _ids_to_results(self, ids):
89 return [
90 self.url_result(vid_id, 'Youtube', video_id=vid_id)
91 for vid_id in ids]
92
93 def _login(self):
94 """
95 Attempt to log in to YouTube.
96 True is returned if successful or skipped.
97 False is returned if login failed.
98
99 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
100 """
101 username, password = self._get_login_info()
102 # No authentication to be performed
103 if username is None:
104 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
105 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
106 if self._downloader.params.get('cookiefile') and False: # TODO remove 'and False' later - too many people using outdated cookies and open issues, remind them.
107 self.to_screen('[Cookies] Reminder - Make sure to always use up to date cookies!')
108 return True
109
110 login_page = self._download_webpage(
111 self._LOGIN_URL, None,
112 note='Downloading login page',
113 errnote='unable to fetch login page', fatal=False)
114 if login_page is False:
115 return
116
117 login_form = self._hidden_inputs(login_page)
118
119 def req(url, f_req, note, errnote):
120 data = login_form.copy()
121 data.update({
122 'pstMsg': 1,
123 'checkConnection': 'youtube',
124 'checkedDomains': 'youtube',
125 'hl': 'en',
126 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
127 'f.req': json.dumps(f_req),
128 'flowName': 'GlifWebSignIn',
129 'flowEntry': 'ServiceLogin',
130 # TODO: reverse actual botguard identifier generation algo
131 'bgRequest': '["identifier",""]',
132 })
133 return self._download_json(
134 url, None, note=note, errnote=errnote,
135 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
136 fatal=False,
137 data=urlencode_postdata(data), headers={
138 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
139 'Google-Accounts-XSRF': 1,
140 })
141
142 def warn(message):
143 self._downloader.report_warning(message)
144
145 lookup_req = [
146 username,
147 None, [], None, 'US', None, None, 2, False, True,
148 [
149 None, None,
150 [2, 1, None, 1,
151 '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',
152 None, [], 4],
153 1, [None, None, []], None, None, None, True
154 ],
155 username,
156 ]
157
158 lookup_results = req(
159 self._LOOKUP_URL, lookup_req,
160 'Looking up account info', 'Unable to look up account info')
161
162 if lookup_results is False:
163 return False
164
165 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
166 if not user_hash:
167 warn('Unable to extract user hash')
168 return False
169
170 challenge_req = [
171 user_hash,
172 None, 1, None, [1, None, None, None, [password, None, True]],
173 [
174 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],
175 1, [None, None, []], None, None, None, True
176 ]]
177
178 challenge_results = req(
179 self._CHALLENGE_URL, challenge_req,
180 'Logging in', 'Unable to log in')
181
182 if challenge_results is False:
183 return
184
185 login_res = try_get(challenge_results, lambda x: x[0][5], list)
186 if login_res:
187 login_msg = try_get(login_res, lambda x: x[5], compat_str)
188 warn(
189 'Unable to login: %s' % 'Invalid password'
190 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
191 return False
192
193 res = try_get(challenge_results, lambda x: x[0][-1], list)
194 if not res:
195 warn('Unable to extract result entry')
196 return False
197
198 login_challenge = try_get(res, lambda x: x[0][0], list)
199 if login_challenge:
200 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
201 if challenge_str == 'TWO_STEP_VERIFICATION':
202 # SEND_SUCCESS - TFA code has been successfully sent to phone
203 # QUOTA_EXCEEDED - reached the limit of TFA codes
204 status = try_get(login_challenge, lambda x: x[5], compat_str)
205 if status == 'QUOTA_EXCEEDED':
206 warn('Exceeded the limit of TFA codes, try later')
207 return False
208
209 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
210 if not tl:
211 warn('Unable to extract TL')
212 return False
213
214 tfa_code = self._get_tfa_info('2-step verification code')
215
216 if not tfa_code:
217 warn(
218 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
219 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
220 return False
221
222 tfa_code = remove_start(tfa_code, 'G-')
223
224 tfa_req = [
225 user_hash, None, 2, None,
226 [
227 9, None, None, None, None, None, None, None,
228 [None, tfa_code, True, 2]
229 ]]
230
231 tfa_results = req(
232 self._TFA_URL.format(tl), tfa_req,
233 'Submitting TFA code', 'Unable to submit TFA code')
234
235 if tfa_results is False:
236 return False
237
238 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
239 if tfa_res:
240 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
241 warn(
242 'Unable to finish TFA: %s' % 'Invalid TFA code'
243 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
244 return False
245
246 check_cookie_url = try_get(
247 tfa_results, lambda x: x[0][-1][2], compat_str)
248 else:
249 CHALLENGES = {
250 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
251 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
252 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
253 }
254 challenge = CHALLENGES.get(
255 challenge_str,
256 '%s returned error %s.' % (self.IE_NAME, challenge_str))
257 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
258 return False
259 else:
260 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
261
262 if not check_cookie_url:
263 warn('Unable to extract CheckCookie URL')
264 return False
265
266 check_cookie_results = self._download_webpage(
267 check_cookie_url, None, 'Checking cookie', fatal=False)
268
269 if check_cookie_results is False:
270 return False
271
272 if 'https://myaccount.google.com/' not in check_cookie_results:
273 warn('Unable to log in')
274 return False
275
276 return True
277
278 def _download_webpage_handle(self, *args, **kwargs):
279 query = kwargs.get('query', {}).copy()
280 kwargs['query'] = query
281 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
282 *args, **compat_kwargs(kwargs))
283
284 def _get_yt_initial_data(self, video_id, webpage):
285 config = self._search_regex(
286 (r'window\["ytInitialData"\]\s*=\s*(.*?)(?<=});',
287 r'var\s+ytInitialData\s*=\s*(.*?)(?<=});'),
288 webpage, 'ytInitialData', default=None)
289 if config:
290 return self._parse_json(
291 uppercase_escape(config), video_id, fatal=False)
292
293 def _real_initialize(self):
294 if self._downloader is None:
295 return
296 self._set_language()
297 if not self._login():
298 return
299
300 _DEFAULT_API_DATA = {
301 'context': {
302 'client': {
303 'clientName': 'WEB',
304 'clientVersion': '2.20201021.03.00',
305 }
306 },
307 }
308
309 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
310
311 def _call_api(self, ep, query, video_id):
312 data = self._DEFAULT_API_DATA.copy()
313 data.update(query)
314
315 response = self._download_json(
316 'https://www.youtube.com/youtubei/v1/%s' % ep, video_id=video_id,
317 note='Downloading API JSON', errnote='Unable to download API page',
318 data=json.dumps(data).encode('utf8'),
319 headers={'content-type': 'application/json'},
320 query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'})
321
322 return response
323
324 def _extract_yt_initial_data(self, video_id, webpage):
325 return self._parse_json(
326 self._search_regex(
327 (r'%s\s*\n' % self._YT_INITIAL_DATA_RE,
328 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
329 video_id)
330
331
332 class YoutubeIE(YoutubeBaseInfoExtractor):
333 IE_DESC = 'YouTube.com'
334 _VALID_URL = r"""(?x)^
335 (
336 (?:https?://|//) # http(s):// or protocol-independent URL
337 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com/|
338 (?:www\.)?deturl\.com/www\.youtube\.com/|
339 (?:www\.)?pwnyoutube\.com/|
340 (?:www\.)?hooktube\.com/|
341 (?:www\.)?yourepeat\.com/|
342 tube\.majestyc\.net/|
343 # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances
344 (?:(?:www|dev)\.)?invidio\.us/|
345 (?:(?:www|no)\.)?invidiou\.sh/|
346 (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/|
347 (?:www\.)?invidious\.kabi\.tk/|
348 (?:www\.)?invidious\.13ad\.de/|
349 (?:www\.)?invidious\.mastodon\.host/|
350 (?:www\.)?invidious\.nixnet\.xyz/|
351 (?:www\.)?invidious\.drycat\.fr/|
352 (?:www\.)?tube\.poal\.co/|
353 (?:www\.)?vid\.wxzm\.sx/|
354 (?:www\.)?yewtu\.be/|
355 (?:www\.)?yt\.elukerio\.org/|
356 (?:www\.)?yt\.lelux\.fi/|
357 (?:www\.)?invidious\.ggc-project\.de/|
358 (?:www\.)?yt\.maisputain\.ovh/|
359 (?:www\.)?invidious\.13ad\.de/|
360 (?:www\.)?invidious\.toot\.koeln/|
361 (?:www\.)?invidious\.fdn\.fr/|
362 (?:www\.)?watch\.nettohikari\.com/|
363 (?:www\.)?kgg2m7yk5aybusll\.onion/|
364 (?:www\.)?qklhadlycap4cnod\.onion/|
365 (?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion/|
366 (?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion/|
367 (?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion/|
368 (?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion/|
369 (?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p/|
370 (?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion/|
371 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
372 (?:.*?\#/)? # handle anchor (#/) redirect urls
373 (?: # the various things that can precede the ID:
374 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
375 |(?: # or the v= param in all its forms
376 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
377 (?:\?|\#!?) # the params delimiter ? or # or #!
378 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
379 v=
380 )
381 ))
382 |(?:
383 youtu\.be| # just youtu.be/xxxx
384 vid\.plus| # or vid.plus/xxxx
385 zwearz\.com/watch| # or zwearz.com/watch/xxxx
386 )/
387 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
388 )
389 )? # all until now is optional -> you can pass the naked ID
390 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
391 (?!.*?\blist=
392 (?:
393 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
394 WL # WL are handled by the watch later IE
395 )
396 )
397 (?(1).+)? # if we found the ID, everything can follow
398 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
399 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
400 _PLAYER_INFO_RE = (
401 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?/base\.(?P<ext>[a-z]+)$',
402 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.(?P<ext>[a-z]+)$',
403 )
404 _formats = {
405 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
406 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
407 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
408 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
409 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
410 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
411 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
412 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
413 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
414 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
415 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
416 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
417 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
418 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
419 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
420 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
421 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
422 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
423
424
425 # 3D videos
426 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
427 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
428 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
429 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
430 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
431 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
432 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
433
434 # Apple HTTP Live Streaming
435 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
436 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
437 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
438 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
439 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
440 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
441 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
442 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
443
444 # DASH mp4 video
445 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
446 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
447 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
448 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
449 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
450 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
451 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
452 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
453 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
454 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
455 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
456 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
457
458 # Dash mp4 audio
459 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
460 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
461 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
462 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
463 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
464 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
465 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
466
467 # Dash webm
468 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
469 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
470 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
471 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
472 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
473 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
474 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
475 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
476 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
477 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
478 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
479 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
480 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
481 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
482 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
483 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
484 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
485 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
486 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
487 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
488 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
489 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
490
491 # Dash webm audio
492 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
493 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
494
495 # Dash webm audio with opus inside
496 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
497 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
498 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
499
500 # RTMP (unnamed)
501 '_rtmp': {'protocol': 'rtmp'},
502
503 # av01 video only formats sometimes served with "unknown" codecs
504 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
505 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
506 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
507 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
508 }
509 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt') # TODO 'json3' raising issues with automatic captions
510
511 _GEO_BYPASS = False
512
513 IE_NAME = 'youtube'
514 _TESTS = [
515 {
516 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
517 'info_dict': {
518 'id': 'BaW_jenozKc',
519 'ext': 'mp4',
520 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
521 'uploader': 'Philipp Hagemeister',
522 'uploader_id': 'phihag',
523 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
524 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
525 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
526 'upload_date': '20121002',
527 '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 .',
528 'categories': ['Science & Technology'],
529 'tags': ['youtube-dl'],
530 'duration': 10,
531 'view_count': int,
532 'like_count': int,
533 'dislike_count': int,
534 'start_time': 1,
535 'end_time': 9,
536 }
537 },
538 {
539 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
540 'note': 'Embed-only video (#1746)',
541 'info_dict': {
542 'id': 'yZIXLfi8CZQ',
543 'ext': 'mp4',
544 'upload_date': '20120608',
545 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
546 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
547 'uploader': 'SET India',
548 'uploader_id': 'setindia',
549 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
550 'age_limit': 18,
551 }
552 },
553 {
554 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
555 'note': 'Use the first video ID in the URL',
556 'info_dict': {
557 'id': 'BaW_jenozKc',
558 'ext': 'mp4',
559 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
560 'uploader': 'Philipp Hagemeister',
561 'uploader_id': 'phihag',
562 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
563 'upload_date': '20121002',
564 '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 .',
565 'categories': ['Science & Technology'],
566 'tags': ['youtube-dl'],
567 'duration': 10,
568 'view_count': int,
569 'like_count': int,
570 'dislike_count': int,
571 },
572 'params': {
573 'skip_download': True,
574 },
575 },
576 {
577 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
578 'note': '256k DASH audio (format 141) via DASH manifest',
579 'info_dict': {
580 'id': 'a9LDPn-MO4I',
581 'ext': 'm4a',
582 'upload_date': '20121002',
583 'uploader_id': '8KVIDEO',
584 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
585 'description': '',
586 'uploader': '8KVIDEO',
587 'title': 'UHDTV TEST 8K VIDEO.mp4'
588 },
589 'params': {
590 'youtube_include_dash_manifest': True,
591 'format': '141',
592 },
593 'skip': 'format 141 not served anymore',
594 },
595 # DASH manifest with encrypted signature
596 {
597 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
598 'info_dict': {
599 'id': 'IB3lcPjvWLA',
600 'ext': 'm4a',
601 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
602 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
603 'duration': 244,
604 'uploader': 'AfrojackVEVO',
605 'uploader_id': 'AfrojackVEVO',
606 'upload_date': '20131011',
607 },
608 'params': {
609 'youtube_include_dash_manifest': True,
610 'format': '141/bestaudio[ext=m4a]',
611 },
612 },
613 # Controversy video
614 {
615 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
616 'info_dict': {
617 'id': 'T4XJQO3qol8',
618 'ext': 'mp4',
619 'duration': 219,
620 'upload_date': '20100909',
621 'uploader': 'Amazing Atheist',
622 'uploader_id': 'TheAmazingAtheist',
623 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
624 'title': 'Burning Everyone\'s Koran',
625 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\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',
626 }
627 },
628 # Normal age-gate video (embed allowed)
629 {
630 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
631 'info_dict': {
632 'id': 'HtVdAasjOgU',
633 'ext': 'mp4',
634 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
635 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
636 'duration': 142,
637 'uploader': 'The Witcher',
638 'uploader_id': 'WitcherGame',
639 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
640 'upload_date': '20140605',
641 'age_limit': 18,
642 },
643 },
644 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
645 # YouTube Red ad is not captured for creator
646 {
647 'url': '__2ABJjxzNo',
648 'info_dict': {
649 'id': '__2ABJjxzNo',
650 'ext': 'mp4',
651 'duration': 266,
652 'upload_date': '20100430',
653 'uploader_id': 'deadmau5',
654 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
655 'creator': 'Dada Life, deadmau5',
656 'description': 'md5:12c56784b8032162bb936a5f76d55360',
657 'uploader': 'deadmau5',
658 'title': 'Deadmau5 - Some Chords (HD)',
659 'alt_title': 'This Machine Kills Some Chords',
660 },
661 'expected_warnings': [
662 'DASH manifest missing',
663 ]
664 },
665 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
666 {
667 'url': 'lqQg6PlCWgI',
668 'info_dict': {
669 'id': 'lqQg6PlCWgI',
670 'ext': 'mp4',
671 'duration': 6085,
672 'upload_date': '20150827',
673 'uploader_id': 'olympic',
674 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
675 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
676 'uploader': 'Olympic',
677 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
678 },
679 'params': {
680 'skip_download': 'requires avconv',
681 }
682 },
683 # Non-square pixels
684 {
685 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
686 'info_dict': {
687 'id': '_b-2C3KPAM0',
688 'ext': 'mp4',
689 'stretched_ratio': 16 / 9.,
690 'duration': 85,
691 'upload_date': '20110310',
692 'uploader_id': 'AllenMeow',
693 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
694 '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',
695 'uploader': '孫ᄋᄅ',
696 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
697 },
698 },
699 # url_encoded_fmt_stream_map is empty string
700 {
701 'url': 'qEJwOuvDf7I',
702 'info_dict': {
703 'id': 'qEJwOuvDf7I',
704 'ext': 'webm',
705 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
706 'description': '',
707 'upload_date': '20150404',
708 'uploader_id': 'spbelect',
709 'uploader': 'Наблюдатели Петербурга',
710 },
711 'params': {
712 'skip_download': 'requires avconv',
713 },
714 'skip': 'This live event has ended.',
715 },
716 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
717 {
718 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
719 'info_dict': {
720 'id': 'FIl7x6_3R5Y',
721 'ext': 'webm',
722 'title': 'md5:7b81415841e02ecd4313668cde88737a',
723 'description': 'md5:116377fd2963b81ec4ce64b542173306',
724 'duration': 220,
725 'upload_date': '20150625',
726 'uploader_id': 'dorappi2000',
727 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
728 'uploader': 'dorappi2000',
729 'formats': 'mincount:31',
730 },
731 'skip': 'not actual anymore',
732 },
733 # DASH manifest with segment_list
734 {
735 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
736 'md5': '8ce563a1d667b599d21064e982ab9e31',
737 'info_dict': {
738 'id': 'CsmdDsKjzN8',
739 'ext': 'mp4',
740 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
741 'uploader': 'Airtek',
742 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
743 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
744 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
745 },
746 'params': {
747 'youtube_include_dash_manifest': True,
748 'format': '135', # bestvideo
749 },
750 'skip': 'This live event has ended.',
751 },
752 {
753 # Multifeed videos (multiple cameras), URL is for Main Camera
754 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
755 'info_dict': {
756 'id': 'jqWvoWXjCVs',
757 'title': 'teamPGP: Rocket League Noob Stream',
758 'description': 'md5:dc7872fb300e143831327f1bae3af010',
759 },
760 'playlist': [{
761 'info_dict': {
762 'id': 'jqWvoWXjCVs',
763 'ext': 'mp4',
764 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
765 'description': 'md5:dc7872fb300e143831327f1bae3af010',
766 'duration': 7335,
767 'upload_date': '20150721',
768 'uploader': 'Beer Games Beer',
769 'uploader_id': 'beergamesbeer',
770 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
771 'license': 'Standard YouTube License',
772 },
773 }, {
774 'info_dict': {
775 'id': '6h8e8xoXJzg',
776 'ext': 'mp4',
777 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
778 'description': 'md5:dc7872fb300e143831327f1bae3af010',
779 'duration': 7337,
780 'upload_date': '20150721',
781 'uploader': 'Beer Games Beer',
782 'uploader_id': 'beergamesbeer',
783 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
784 'license': 'Standard YouTube License',
785 },
786 }, {
787 'info_dict': {
788 'id': 'PUOgX5z9xZw',
789 'ext': 'mp4',
790 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
791 'description': 'md5:dc7872fb300e143831327f1bae3af010',
792 'duration': 7337,
793 'upload_date': '20150721',
794 'uploader': 'Beer Games Beer',
795 'uploader_id': 'beergamesbeer',
796 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
797 'license': 'Standard YouTube License',
798 },
799 }, {
800 'info_dict': {
801 'id': 'teuwxikvS5k',
802 'ext': 'mp4',
803 'title': 'teamPGP: Rocket League Noob Stream (zim)',
804 'description': 'md5:dc7872fb300e143831327f1bae3af010',
805 'duration': 7334,
806 'upload_date': '20150721',
807 'uploader': 'Beer Games Beer',
808 'uploader_id': 'beergamesbeer',
809 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
810 'license': 'Standard YouTube License',
811 },
812 }],
813 'params': {
814 'skip_download': True,
815 },
816 'skip': 'This video is not available.',
817 },
818 {
819 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
820 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
821 'info_dict': {
822 'id': 'gVfLd0zydlo',
823 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
824 },
825 'playlist_count': 2,
826 'skip': 'Not multifeed anymore',
827 },
828 {
829 'url': 'https://vid.plus/FlRa-iH7PGw',
830 'only_matching': True,
831 },
832 {
833 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
834 'only_matching': True,
835 },
836 {
837 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
838 # Also tests cut-off URL expansion in video description (see
839 # https://github.com/ytdl-org/youtube-dl/issues/1892,
840 # https://github.com/ytdl-org/youtube-dl/issues/8164)
841 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
842 'info_dict': {
843 'id': 'lsguqyKfVQg',
844 'ext': 'mp4',
845 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
846 'alt_title': 'Dark Walk - Position Music',
847 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
848 'duration': 133,
849 'upload_date': '20151119',
850 'uploader_id': 'IronSoulElf',
851 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
852 'uploader': 'IronSoulElf',
853 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
854 'track': 'Dark Walk - Position Music',
855 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
856 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
857 },
858 'params': {
859 'skip_download': True,
860 },
861 },
862 {
863 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
864 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
865 'only_matching': True,
866 },
867 {
868 # Video with yt:stretch=17:0
869 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
870 'info_dict': {
871 'id': 'Q39EVAstoRM',
872 'ext': 'mp4',
873 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
874 'description': 'md5:ee18a25c350637c8faff806845bddee9',
875 'upload_date': '20151107',
876 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
877 'uploader': 'CH GAMER DROID',
878 },
879 'params': {
880 'skip_download': True,
881 },
882 'skip': 'This video does not exist.',
883 },
884 {
885 # Video licensed under Creative Commons
886 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
887 'info_dict': {
888 'id': 'M4gD1WSo5mA',
889 'ext': 'mp4',
890 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
891 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
892 'duration': 721,
893 'upload_date': '20150127',
894 'uploader_id': 'BerkmanCenter',
895 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
896 'uploader': 'The Berkman Klein Center for Internet & Society',
897 'license': 'Creative Commons Attribution license (reuse allowed)',
898 },
899 'params': {
900 'skip_download': True,
901 },
902 },
903 {
904 # Channel-like uploader_url
905 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
906 'info_dict': {
907 'id': 'eQcmzGIKrzg',
908 'ext': 'mp4',
909 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
910 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
911 'duration': 4060,
912 'upload_date': '20151119',
913 'uploader': 'Bernie Sanders',
914 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
915 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
916 'license': 'Creative Commons Attribution license (reuse allowed)',
917 },
918 'params': {
919 'skip_download': True,
920 },
921 },
922 {
923 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
924 'only_matching': True,
925 },
926 {
927 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
928 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
929 'only_matching': True,
930 },
931 {
932 # Rental video preview
933 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
934 'info_dict': {
935 'id': 'uGpuVWrhIzE',
936 'ext': 'mp4',
937 'title': 'Piku - Trailer',
938 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
939 'upload_date': '20150811',
940 'uploader': 'FlixMatrix',
941 'uploader_id': 'FlixMatrixKaravan',
942 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
943 'license': 'Standard YouTube License',
944 },
945 'params': {
946 'skip_download': True,
947 },
948 'skip': 'This video is not available.',
949 },
950 {
951 # YouTube Red video with episode data
952 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
953 'info_dict': {
954 'id': 'iqKdEhx-dD4',
955 'ext': 'mp4',
956 'title': 'Isolation - Mind Field (Ep 1)',
957 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
958 'duration': 2085,
959 'upload_date': '20170118',
960 'uploader': 'Vsauce',
961 'uploader_id': 'Vsauce',
962 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
963 'series': 'Mind Field',
964 'season_number': 1,
965 'episode_number': 1,
966 },
967 'params': {
968 'skip_download': True,
969 },
970 'expected_warnings': [
971 'Skipping DASH manifest',
972 ],
973 },
974 {
975 # The following content has been identified by the YouTube community
976 # as inappropriate or offensive to some audiences.
977 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
978 'info_dict': {
979 'id': '6SJNVb0GnPI',
980 'ext': 'mp4',
981 'title': 'Race Differences in Intelligence',
982 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
983 'duration': 965,
984 'upload_date': '20140124',
985 'uploader': 'New Century Foundation',
986 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
987 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
988 },
989 'params': {
990 'skip_download': True,
991 },
992 },
993 {
994 # itag 212
995 'url': '1t24XAntNCY',
996 'only_matching': True,
997 },
998 {
999 # geo restricted to JP
1000 'url': 'sJL6WA-aGkQ',
1001 'only_matching': True,
1002 },
1003 {
1004 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1005 'only_matching': True,
1006 },
1007 {
1008 # DRM protected
1009 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1010 'only_matching': True,
1011 },
1012 {
1013 # Video with unsupported adaptive stream type formats
1014 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1015 'info_dict': {
1016 'id': 'Z4Vy8R84T1U',
1017 'ext': 'mp4',
1018 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1019 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1020 'duration': 433,
1021 'upload_date': '20130923',
1022 'uploader': 'Amelia Putri Harwita',
1023 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1024 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1025 'formats': 'maxcount:10',
1026 },
1027 'params': {
1028 'skip_download': True,
1029 'youtube_include_dash_manifest': False,
1030 },
1031 'skip': 'not actual anymore',
1032 },
1033 {
1034 # Youtube Music Auto-generated description
1035 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1036 'info_dict': {
1037 'id': 'MgNrAu2pzNs',
1038 'ext': 'mp4',
1039 'title': 'Voyeur Girl',
1040 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1041 'upload_date': '20190312',
1042 'uploader': 'Stephen - Topic',
1043 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1044 'artist': 'Stephen',
1045 'track': 'Voyeur Girl',
1046 'album': 'it\'s too much love to know my dear',
1047 'release_date': '20190313',
1048 'release_year': 2019,
1049 },
1050 'params': {
1051 'skip_download': True,
1052 },
1053 },
1054 {
1055 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1056 'only_matching': True,
1057 },
1058 {
1059 # invalid -> valid video id redirection
1060 'url': 'DJztXj2GPfl',
1061 'info_dict': {
1062 'id': 'DJztXj2GPfk',
1063 'ext': 'mp4',
1064 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1065 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1066 'upload_date': '20090125',
1067 'uploader': 'Prochorowka',
1068 'uploader_id': 'Prochorowka',
1069 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1070 'artist': 'Panjabi MC',
1071 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1072 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1073 },
1074 'params': {
1075 'skip_download': True,
1076 },
1077 },
1078 {
1079 # empty description results in an empty string
1080 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1081 'info_dict': {
1082 'id': 'x41yOUIvK2k',
1083 'ext': 'mp4',
1084 'title': 'IMG 3456',
1085 'description': '',
1086 'upload_date': '20170613',
1087 'uploader_id': 'ElevageOrVert',
1088 'uploader': 'ElevageOrVert',
1089 },
1090 'params': {
1091 'skip_download': True,
1092 },
1093 },
1094 {
1095 # with '};' inside yt initial data (see https://github.com/ytdl-org/youtube-dl/issues/27093)
1096 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1097 'info_dict': {
1098 'id': 'CHqg6qOn4no',
1099 'ext': 'mp4',
1100 'title': 'Part 77 Sort a list of simple types in c#',
1101 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1102 'upload_date': '20130831',
1103 'uploader_id': 'kudvenkat',
1104 'uploader': 'kudvenkat',
1105 },
1106 'params': {
1107 'skip_download': True,
1108 },
1109 },
1110 ]
1111
1112 def __init__(self, *args, **kwargs):
1113 super(YoutubeIE, self).__init__(*args, **kwargs)
1114 self._player_cache = {}
1115
1116 def report_video_info_webpage_download(self, video_id):
1117 """Report attempt to download video info webpage."""
1118 self.to_screen('%s: Downloading video info webpage' % video_id)
1119
1120 def report_information_extraction(self, video_id):
1121 """Report attempt to extract video information."""
1122 self.to_screen('%s: Extracting video information' % video_id)
1123
1124 def report_unavailable_format(self, video_id, format):
1125 """Report extracted video URL."""
1126 self.to_screen('%s: Format %s not available' % (video_id, format))
1127
1128 def report_rtmp_download(self):
1129 """Indicate the download will use the RTMP protocol."""
1130 self.to_screen('RTMP download detected')
1131
1132 def _signature_cache_id(self, example_sig):
1133 """ Return a string representation of a signature """
1134 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1135
1136 @classmethod
1137 def _extract_player_info(cls, player_url):
1138 for player_re in cls._PLAYER_INFO_RE:
1139 id_m = re.search(player_re, player_url)
1140 if id_m:
1141 break
1142 else:
1143 raise ExtractorError('Cannot identify player %r' % player_url)
1144 return id_m.group('ext'), id_m.group('id')
1145
1146 def _extract_signature_function(self, video_id, player_url, example_sig):
1147 player_type, player_id = self._extract_player_info(player_url)
1148
1149 # Read from filesystem cache
1150 func_id = '%s_%s_%s' % (
1151 player_type, player_id, self._signature_cache_id(example_sig))
1152 assert os.path.basename(func_id) == func_id
1153
1154 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1155 if cache_spec is not None:
1156 return lambda s: ''.join(s[i] for i in cache_spec)
1157
1158 download_note = (
1159 'Downloading player %s' % player_url
1160 if self._downloader.params.get('verbose') else
1161 'Downloading %s player %s' % (player_type, player_id)
1162 )
1163 if player_type == 'js':
1164 code = self._download_webpage(
1165 player_url, video_id,
1166 note=download_note,
1167 errnote='Download of %s failed' % player_url)
1168 res = self._parse_sig_js(code)
1169 elif player_type == 'swf':
1170 urlh = self._request_webpage(
1171 player_url, video_id,
1172 note=download_note,
1173 errnote='Download of %s failed' % player_url)
1174 code = urlh.read()
1175 res = self._parse_sig_swf(code)
1176 else:
1177 assert False, 'Invalid player type %r' % player_type
1178
1179 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1180 cache_res = res(test_string)
1181 cache_spec = [ord(c) for c in cache_res]
1182
1183 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1184 return res
1185
1186 def _print_sig_code(self, func, example_sig):
1187 def gen_sig_code(idxs):
1188 def _genslice(start, end, step):
1189 starts = '' if start == 0 else str(start)
1190 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1191 steps = '' if step == 1 else (':%d' % step)
1192 return 's[%s%s%s]' % (starts, ends, steps)
1193
1194 step = None
1195 # Quelch pyflakes warnings - start will be set when step is set
1196 start = '(Never used)'
1197 for i, prev in zip(idxs[1:], idxs[:-1]):
1198 if step is not None:
1199 if i - prev == step:
1200 continue
1201 yield _genslice(start, prev, step)
1202 step = None
1203 continue
1204 if i - prev in [-1, 1]:
1205 step = i - prev
1206 start = prev
1207 continue
1208 else:
1209 yield 's[%d]' % prev
1210 if step is None:
1211 yield 's[%d]' % i
1212 else:
1213 yield _genslice(start, i, step)
1214
1215 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1216 cache_res = func(test_string)
1217 cache_spec = [ord(c) for c in cache_res]
1218 expr_code = ' + '.join(gen_sig_code(cache_spec))
1219 signature_id_tuple = '(%s)' % (
1220 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1221 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1222 ' return %s\n') % (signature_id_tuple, expr_code)
1223 self.to_screen('Extracted signature function:\n' + code)
1224
1225 def _parse_sig_js(self, jscode):
1226 funcname = self._search_regex(
1227 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1228 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1229 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\(\s*""\s*\)',
1230 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1231 # Obsolete patterns
1232 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1233 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1234 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1235 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1236 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1237 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1238 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1239 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1240 jscode, 'Initial JS player signature function name', group='sig')
1241
1242 jsi = JSInterpreter(jscode)
1243 initial_function = jsi.extract_function(funcname)
1244 return lambda s: initial_function([s])
1245
1246 def _parse_sig_swf(self, file_contents):
1247 swfi = SWFInterpreter(file_contents)
1248 TARGET_CLASSNAME = 'SignatureDecipher'
1249 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1250 initial_function = swfi.extract_function(searched_class, 'decipher')
1251 return lambda s: initial_function([s])
1252
1253 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1254 """Turn the encrypted s field into a working signature"""
1255
1256 if player_url is None:
1257 raise ExtractorError('Cannot decrypt signature without player_url')
1258
1259 if player_url.startswith('//'):
1260 player_url = 'https:' + player_url
1261 elif not re.match(r'https?://', player_url):
1262 player_url = compat_urlparse.urljoin(
1263 'https://www.youtube.com', player_url)
1264 try:
1265 player_id = (player_url, self._signature_cache_id(s))
1266 if player_id not in self._player_cache:
1267 func = self._extract_signature_function(
1268 video_id, player_url, s
1269 )
1270 self._player_cache[player_id] = func
1271 func = self._player_cache[player_id]
1272 if self._downloader.params.get('youtube_print_sig_code'):
1273 self._print_sig_code(func, s)
1274 return func(s)
1275 except Exception as e:
1276 tb = traceback.format_exc()
1277 raise ExtractorError(
1278 'Signature extraction failed: ' + tb, cause=e)
1279
1280 def _get_subtitles(self, video_id, webpage, has_live_chat_replay):
1281 try:
1282 subs_doc = self._download_xml(
1283 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1284 video_id, note=False)
1285 except ExtractorError as err:
1286 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1287 return {}
1288
1289 sub_lang_list = {}
1290 for track in subs_doc.findall('track'):
1291 lang = track.attrib['lang_code']
1292 if lang in sub_lang_list:
1293 continue
1294 sub_formats = []
1295 for ext in self._SUBTITLE_FORMATS:
1296 params = compat_urllib_parse_urlencode({
1297 'lang': lang,
1298 'v': video_id,
1299 'fmt': ext,
1300 'name': track.attrib['name'].encode('utf-8'),
1301 })
1302 sub_formats.append({
1303 'url': 'https://www.youtube.com/api/timedtext?' + params,
1304 'ext': ext,
1305 })
1306 sub_lang_list[lang] = sub_formats
1307 if has_live_chat_replay:
1308 sub_lang_list['live_chat'] = [
1309 {
1310 'video_id': video_id,
1311 'ext': 'json',
1312 'protocol': 'youtube_live_chat_replay',
1313 },
1314 ]
1315 if not sub_lang_list:
1316 self._downloader.report_warning('video doesn\'t have subtitles')
1317 return {}
1318 return sub_lang_list
1319
1320 def _get_ytplayer_config(self, video_id, webpage):
1321 patterns = (
1322 # User data may contain arbitrary character sequences that may affect
1323 # JSON extraction with regex, e.g. when '};' is contained the second
1324 # regex won't capture the whole JSON. Yet working around by trying more
1325 # concrete regex first keeping in mind proper quoted string handling
1326 # to be implemented in future that will replace this workaround (see
1327 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1328 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1329 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1330 r';ytplayer\.config\s*=\s*({.+?});',
1331 )
1332 config = self._search_regex(
1333 patterns, webpage, 'ytplayer.config', default=None)
1334 if config:
1335 return self._parse_json(
1336 uppercase_escape(config), video_id, fatal=False)
1337
1338 def _get_automatic_captions(self, video_id, webpage):
1339 """We need the webpage for getting the captions url, pass it as an
1340 argument to speed up the process."""
1341 self.to_screen('%s: Looking for automatic captions' % video_id)
1342 player_config = self._get_ytplayer_config(video_id, webpage)
1343 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1344 if not player_config:
1345 self._downloader.report_warning(err_msg)
1346 return {}
1347 try:
1348 args = player_config['args']
1349 caption_url = args.get('ttsurl')
1350 if caption_url:
1351 timestamp = args['timestamp']
1352 # We get the available subtitles
1353 list_params = compat_urllib_parse_urlencode({
1354 'type': 'list',
1355 'tlangs': 1,
1356 'asrs': 1,
1357 })
1358 list_url = caption_url + '&' + list_params
1359 caption_list = self._download_xml(list_url, video_id)
1360 original_lang_node = caption_list.find('track')
1361 if original_lang_node is None:
1362 self._downloader.report_warning('Video doesn\'t have automatic captions')
1363 return {}
1364 original_lang = original_lang_node.attrib['lang_code']
1365 caption_kind = original_lang_node.attrib.get('kind', '')
1366
1367 sub_lang_list = {}
1368 for lang_node in caption_list.findall('target'):
1369 sub_lang = lang_node.attrib['lang_code']
1370 sub_formats = []
1371 for ext in self._SUBTITLE_FORMATS:
1372 params = compat_urllib_parse_urlencode({
1373 'lang': original_lang,
1374 'tlang': sub_lang,
1375 'fmt': ext,
1376 'ts': timestamp,
1377 'kind': caption_kind,
1378 })
1379 sub_formats.append({
1380 'url': caption_url + '&' + params,
1381 'ext': ext,
1382 })
1383 sub_lang_list[sub_lang] = sub_formats
1384 return sub_lang_list
1385
1386 def make_captions(sub_url, sub_langs):
1387 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1388 caption_qs = compat_parse_qs(parsed_sub_url.query)
1389 captions = {}
1390 for sub_lang in sub_langs:
1391 sub_formats = []
1392 for ext in self._SUBTITLE_FORMATS:
1393 caption_qs.update({
1394 'tlang': [sub_lang],
1395 'fmt': [ext],
1396 })
1397 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1398 query=compat_urllib_parse_urlencode(caption_qs, True)))
1399 sub_formats.append({
1400 'url': sub_url,
1401 'ext': ext,
1402 })
1403 captions[sub_lang] = sub_formats
1404 return captions
1405
1406 # New captions format as of 22.06.2017
1407 player_response = args.get('player_response')
1408 if player_response and isinstance(player_response, compat_str):
1409 player_response = self._parse_json(
1410 player_response, video_id, fatal=False)
1411 if player_response:
1412 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1413 base_url = renderer['captionTracks'][0]['baseUrl']
1414 sub_lang_list = []
1415 for lang in renderer['translationLanguages']:
1416 lang_code = lang.get('languageCode')
1417 if lang_code:
1418 sub_lang_list.append(lang_code)
1419 return make_captions(base_url, sub_lang_list)
1420
1421 # Some videos don't provide ttsurl but rather caption_tracks and
1422 # caption_translation_languages (e.g. 20LmZk1hakA)
1423 # Does not used anymore as of 22.06.2017
1424 caption_tracks = args['caption_tracks']
1425 caption_translation_languages = args['caption_translation_languages']
1426 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1427 sub_lang_list = []
1428 for lang in caption_translation_languages.split(','):
1429 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1430 sub_lang = lang_qs.get('lc', [None])[0]
1431 if sub_lang:
1432 sub_lang_list.append(sub_lang)
1433 return make_captions(caption_url, sub_lang_list)
1434 # An extractor error can be raise by the download process if there are
1435 # no automatic captions but there are subtitles
1436 except (KeyError, IndexError, ExtractorError):
1437 self._downloader.report_warning(err_msg)
1438 return {}
1439
1440 def _mark_watched(self, video_id, video_info, player_response):
1441 playback_url = url_or_none(try_get(
1442 player_response,
1443 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1444 video_info, lambda x: x['videostats_playback_base_url'][0]))
1445 if not playback_url:
1446 return
1447 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1448 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1449
1450 # cpn generation algorithm is reverse engineered from base.js.
1451 # In fact it works even with dummy cpn.
1452 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1453 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1454
1455 qs.update({
1456 'ver': ['2'],
1457 'cpn': [cpn],
1458 })
1459 playback_url = compat_urlparse.urlunparse(
1460 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1461
1462 self._download_webpage(
1463 playback_url, video_id, 'Marking watched',
1464 'Unable to mark watched', fatal=False)
1465
1466 @staticmethod
1467 def _extract_urls(webpage):
1468 # Embedded YouTube player
1469 entries = [
1470 unescapeHTML(mobj.group('url'))
1471 for mobj in re.finditer(r'''(?x)
1472 (?:
1473 <iframe[^>]+?src=|
1474 data-video-url=|
1475 <embed[^>]+?src=|
1476 embedSWF\(?:\s*|
1477 <object[^>]+data=|
1478 new\s+SWFObject\(
1479 )
1480 (["\'])
1481 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1482 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1483 \1''', webpage)]
1484
1485 # lazyYT YouTube embed
1486 entries.extend(list(map(
1487 unescapeHTML,
1488 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1489
1490 # Wordpress "YouTube Video Importer" plugin
1491 matches = re.findall(r'''(?x)<div[^>]+
1492 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1493 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1494 entries.extend(m[-1] for m in matches)
1495
1496 return entries
1497
1498 @staticmethod
1499 def _extract_url(webpage):
1500 urls = YoutubeIE._extract_urls(webpage)
1501 return urls[0] if urls else None
1502
1503 @classmethod
1504 def extract_id(cls, url):
1505 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1506 if mobj is None:
1507 raise ExtractorError('Invalid URL: %s' % url)
1508 video_id = mobj.group(2)
1509 return video_id
1510
1511 def _extract_chapters_from_json(self, webpage, video_id, duration):
1512 if not webpage:
1513 return
1514 data = self._extract_yt_initial_data(video_id, webpage)
1515 if not data or not isinstance(data, dict):
1516 return
1517 chapters_list = try_get(
1518 data,
1519 lambda x: x['playerOverlays']
1520 ['playerOverlayRenderer']
1521 ['decoratedPlayerBarRenderer']
1522 ['decoratedPlayerBarRenderer']
1523 ['playerBar']
1524 ['chapteredPlayerBarRenderer']
1525 ['chapters'],
1526 list)
1527 if not chapters_list:
1528 return
1529
1530 def chapter_time(chapter):
1531 return float_or_none(
1532 try_get(
1533 chapter,
1534 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1535 int),
1536 scale=1000)
1537 chapters = []
1538 for next_num, chapter in enumerate(chapters_list, start=1):
1539 start_time = chapter_time(chapter)
1540 if start_time is None:
1541 continue
1542 end_time = (chapter_time(chapters_list[next_num])
1543 if next_num < len(chapters_list) else duration)
1544 if end_time is None:
1545 continue
1546 title = try_get(
1547 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1548 compat_str)
1549 chapters.append({
1550 'start_time': start_time,
1551 'end_time': end_time,
1552 'title': title,
1553 })
1554 return chapters
1555
1556 @staticmethod
1557 def _extract_chapters_from_description(description, duration):
1558 if not description:
1559 return None
1560 chapter_lines = re.findall(
1561 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1562 description)
1563 if not chapter_lines:
1564 return None
1565 chapters = []
1566 for next_num, (chapter_line, time_point) in enumerate(
1567 chapter_lines, start=1):
1568 start_time = parse_duration(time_point)
1569 if start_time is None:
1570 continue
1571 if start_time > duration:
1572 break
1573 end_time = (duration if next_num == len(chapter_lines)
1574 else parse_duration(chapter_lines[next_num][1]))
1575 if end_time is None:
1576 continue
1577 if end_time > duration:
1578 end_time = duration
1579 if start_time > end_time:
1580 break
1581 chapter_title = re.sub(
1582 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1583 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1584 chapters.append({
1585 'start_time': start_time,
1586 'end_time': end_time,
1587 'title': chapter_title,
1588 })
1589 return chapters
1590
1591 def _extract_chapters(self, webpage, description, video_id, duration):
1592 return (self._extract_chapters_from_json(webpage, video_id, duration)
1593 or self._extract_chapters_from_description(description, duration))
1594
1595 def _real_extract(self, url):
1596 url, smuggled_data = unsmuggle_url(url, {})
1597
1598 proto = (
1599 'http' if self._downloader.params.get('prefer_insecure', False)
1600 else 'https')
1601
1602 start_time = None
1603 end_time = None
1604 parsed_url = compat_urllib_parse_urlparse(url)
1605 for component in [parsed_url.fragment, parsed_url.query]:
1606 query = compat_parse_qs(component)
1607 if start_time is None and 't' in query:
1608 start_time = parse_duration(query['t'][0])
1609 if start_time is None and 'start' in query:
1610 start_time = parse_duration(query['start'][0])
1611 if end_time is None and 'end' in query:
1612 end_time = parse_duration(query['end'][0])
1613
1614 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1615 mobj = re.search(self._NEXT_URL_RE, url)
1616 if mobj:
1617 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1618 video_id = self.extract_id(url)
1619
1620 # Get video webpage
1621 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1622 video_webpage, urlh = self._download_webpage_handle(url, video_id)
1623
1624 qs = compat_parse_qs(compat_urllib_parse_urlparse(urlh.geturl()).query)
1625 video_id = qs.get('v', [None])[0] or video_id
1626
1627 # Attempt to extract SWF player URL
1628 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1629 if mobj is not None:
1630 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1631 else:
1632 player_url = None
1633
1634 dash_mpds = []
1635
1636 def add_dash_mpd(video_info):
1637 dash_mpd = video_info.get('dashmpd')
1638 if dash_mpd and dash_mpd[0] not in dash_mpds:
1639 dash_mpds.append(dash_mpd[0])
1640
1641 def add_dash_mpd_pr(pl_response):
1642 dash_mpd = url_or_none(try_get(
1643 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1644 compat_str))
1645 if dash_mpd and dash_mpd not in dash_mpds:
1646 dash_mpds.append(dash_mpd)
1647
1648 is_live = None
1649 view_count = None
1650
1651 def extract_view_count(v_info):
1652 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1653
1654 def extract_player_response(player_response, video_id):
1655 pl_response = str_or_none(player_response)
1656 if not pl_response:
1657 return
1658 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1659 if isinstance(pl_response, dict):
1660 add_dash_mpd_pr(pl_response)
1661 return pl_response
1662
1663 def extract_embedded_config(embed_webpage, video_id):
1664 embedded_config = self._search_regex(
1665 r'setConfig\(({.*})\);',
1666 embed_webpage, 'ytInitialData', default=None)
1667 if embedded_config:
1668 return embedded_config
1669
1670 player_response = {}
1671
1672 # Get video info
1673 video_info = {}
1674 embed_webpage = None
1675 if (self._og_search_property('restrictions:age', video_webpage, default=None) == '18+'
1676 or re.search(r'player-age-gate-content">', video_webpage) is not None):
1677 cookie_keys = self._get_cookies('https://www.youtube.com').keys()
1678 age_gate = True
1679 # We simulate the access to the video from www.youtube.com/v/{video_id}
1680 # this can be viewed without login into Youtube
1681 url = proto + '://www.youtube.com/embed/%s' % video_id
1682 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1683 ext = extract_embedded_config(embed_webpage, video_id)
1684 # playabilityStatus = re.search(r'{\\\"status\\\":\\\"(?P<playabilityStatus>[^\"]+)\\\"', ext)
1685 playable_in_embed = re.search(r'{\\\"playableInEmbed\\\":(?P<playableinEmbed>[^\,]+)', ext)
1686 if not playable_in_embed:
1687 self.to_screen('Could not determine whether playabale in embed for video %s' % video_id)
1688 playable_in_embed = ''
1689 else:
1690 playable_in_embed = playable_in_embed.group('playableinEmbed')
1691 # check if video is only playable on youtube in other words not playable in embed - if so it requires auth (cookies)
1692 # if re.search(r'player-unavailable">', embed_webpage) is not None:
1693 if playable_in_embed == 'false':
1694 '''
1695 # TODO apply this patch when Support for Python 2.6(!) and above drops
1696 if ({'VISITOR_INFO1_LIVE', 'HSID', 'SSID', 'SID'} <= cookie_keys
1697 or {'VISITOR_INFO1_LIVE', '__Secure-3PSID', 'LOGIN_INFO'} <= cookie_keys):
1698 '''
1699 if (set(('VISITOR_INFO1_LIVE', 'HSID', 'SSID', 'SID')) <= set(cookie_keys)
1700 or set(('VISITOR_INFO1_LIVE', '__Secure-3PSID', 'LOGIN_INFO')) <= set(cookie_keys)):
1701 age_gate = False
1702 # Try looking directly into the video webpage
1703 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1704 if ytplayer_config:
1705 args = ytplayer_config.get("args")
1706 if args is not None:
1707 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1708 # Convert to the same format returned by compat_parse_qs
1709 video_info = dict((k, [v]) for k, v in args.items())
1710 add_dash_mpd(video_info)
1711 # Rental video is not rented but preview is available (e.g.
1712 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1713 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1714 if not video_info and args.get('ypc_vid'):
1715 return self.url_result(
1716 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1717 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1718 is_live = True
1719 if not player_response:
1720 player_response = extract_player_response(args.get('player_response'), video_id)
1721 elif not player_response:
1722 player_response = ytplayer_config
1723 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1724 add_dash_mpd_pr(player_response)
1725 else:
1726 raise ExtractorError('Video is age restricted and only playable on Youtube. Requires cookies!', expected=True)
1727 else:
1728 data = compat_urllib_parse_urlencode({
1729 'video_id': video_id,
1730 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1731 'sts': self._search_regex(
1732 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1733 })
1734 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1735 try:
1736 video_info_webpage = self._download_webpage(
1737 video_info_url, video_id,
1738 note='Refetching age-gated info webpage',
1739 errnote='unable to download video info webpage')
1740 except ExtractorError:
1741 video_info_webpage = None
1742 if video_info_webpage:
1743 video_info = compat_parse_qs(video_info_webpage)
1744 pl_response = video_info.get('player_response', [None])[0]
1745 player_response = extract_player_response(pl_response, video_id)
1746 add_dash_mpd(video_info)
1747 view_count = extract_view_count(video_info)
1748 else:
1749 age_gate = False
1750 # Try looking directly into the video webpage
1751 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1752 if ytplayer_config:
1753 args = ytplayer_config.get('args', {})
1754 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1755 # Convert to the same format returned by compat_parse_qs
1756 video_info = dict((k, [v]) for k, v in args.items())
1757 add_dash_mpd(video_info)
1758 # Rental video is not rented but preview is available (e.g.
1759 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1760 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1761 if not video_info and args.get('ypc_vid'):
1762 return self.url_result(
1763 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1764 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1765 is_live = True
1766 if not player_response:
1767 player_response = extract_player_response(args.get('player_response'), video_id)
1768 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1769 add_dash_mpd_pr(player_response)
1770
1771 if not video_info and not player_response:
1772 player_response = extract_player_response(
1773 self._search_regex(
1774 r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;', video_webpage,
1775 'initial player response', default='{}'),
1776 video_id)
1777
1778 def extract_unavailable_message():
1779 messages = []
1780 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1781 msg = self._html_search_regex(
1782 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1783 video_webpage, 'unavailable %s' % kind, default=None)
1784 if msg:
1785 messages.append(msg)
1786 if messages:
1787 return '\n'.join(messages)
1788
1789 if not video_info and not player_response:
1790 unavailable_message = extract_unavailable_message()
1791 if not unavailable_message:
1792 unavailable_message = 'Unable to extract video data'
1793 raise ExtractorError(
1794 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1795
1796 if not isinstance(video_info, dict):
1797 video_info = {}
1798
1799 video_details = try_get(
1800 player_response, lambda x: x['videoDetails'], dict) or {}
1801
1802 microformat = try_get(
1803 player_response, lambda x: x['microformat']['playerMicroformatRenderer'], dict) or {}
1804
1805 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1806 if not video_title:
1807 self._downloader.report_warning('Unable to extract video title')
1808 video_title = '_'
1809
1810 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1811 if video_description:
1812
1813 def replace_url(m):
1814 redir_url = compat_urlparse.urljoin(url, m.group(1))
1815 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1816 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1817 qs = compat_parse_qs(parsed_redir_url.query)
1818 q = qs.get('q')
1819 if q and q[0]:
1820 return q[0]
1821 return redir_url
1822
1823 description_original = video_description = re.sub(r'''(?x)
1824 <a\s+
1825 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1826 (?:title|href)="([^"]+)"\s+
1827 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1828 class="[^"]*"[^>]*>
1829 [^<]+\.{3}\s*
1830 </a>
1831 ''', replace_url, video_description)
1832 video_description = clean_html(video_description)
1833 else:
1834 video_description = video_details.get('shortDescription')
1835 if video_description is None:
1836 video_description = self._html_search_meta('description', video_webpage)
1837
1838 if not smuggled_data.get('force_singlefeed', False):
1839 if not self._downloader.params.get('noplaylist'):
1840 multifeed_metadata_list = try_get(
1841 player_response,
1842 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1843 compat_str) or try_get(
1844 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1845 if multifeed_metadata_list:
1846 entries = []
1847 feed_ids = []
1848 for feed in multifeed_metadata_list.split(','):
1849 # Unquote should take place before split on comma (,) since textual
1850 # fields may contain comma as well (see
1851 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1852 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1853
1854 def feed_entry(name):
1855 return try_get(feed_data, lambda x: x[name][0], compat_str)
1856
1857 feed_id = feed_entry('id')
1858 if not feed_id:
1859 continue
1860 feed_title = feed_entry('title')
1861 title = video_title
1862 if feed_title:
1863 title += ' (%s)' % feed_title
1864 entries.append({
1865 '_type': 'url_transparent',
1866 'ie_key': 'Youtube',
1867 'url': smuggle_url(
1868 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1869 {'force_singlefeed': True}),
1870 'title': title,
1871 })
1872 feed_ids.append(feed_id)
1873 self.to_screen(
1874 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1875 % (', '.join(feed_ids), video_id))
1876 return self.playlist_result(entries, video_id, video_title, video_description)
1877 else:
1878 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1879
1880 if view_count is None:
1881 view_count = extract_view_count(video_info)
1882 if view_count is None and video_details:
1883 view_count = int_or_none(video_details.get('viewCount'))
1884 if view_count is None and microformat:
1885 view_count = int_or_none(microformat.get('viewCount'))
1886
1887 if is_live is None:
1888 is_live = bool_or_none(video_details.get('isLive'))
1889
1890 has_live_chat_replay = False
1891 if not is_live:
1892 yt_initial_data = self._get_yt_initial_data(video_id, video_webpage)
1893 try:
1894 yt_initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
1895 has_live_chat_replay = True
1896 except (KeyError, IndexError, TypeError):
1897 pass
1898
1899 # Check for "rental" videos
1900 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1901 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1902
1903 def _extract_filesize(media_url):
1904 return int_or_none(self._search_regex(
1905 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1906
1907 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or []
1908 streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or [])
1909
1910 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1911 self.report_rtmp_download()
1912 formats = [{
1913 'format_id': '_rtmp',
1914 'protocol': 'rtmp',
1915 'url': video_info['conn'][0],
1916 'player_url': player_url,
1917 }]
1918 elif not is_live and (streaming_formats or len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1919 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1920 if 'rtmpe%3Dyes' in encoded_url_map:
1921 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1922 formats = []
1923 formats_spec = {}
1924 fmt_list = video_info.get('fmt_list', [''])[0]
1925 if fmt_list:
1926 for fmt in fmt_list.split(','):
1927 spec = fmt.split('/')
1928 if len(spec) > 1:
1929 width_height = spec[1].split('x')
1930 if len(width_height) == 2:
1931 formats_spec[spec[0]] = {
1932 'resolution': spec[1],
1933 'width': int_or_none(width_height[0]),
1934 'height': int_or_none(width_height[1]),
1935 }
1936 for fmt in streaming_formats:
1937 itag = str_or_none(fmt.get('itag'))
1938 if not itag:
1939 continue
1940 quality = fmt.get('quality')
1941 quality_label = fmt.get('qualityLabel') or quality
1942 formats_spec[itag] = {
1943 'asr': int_or_none(fmt.get('audioSampleRate')),
1944 'filesize': int_or_none(fmt.get('contentLength')),
1945 'format_note': quality_label,
1946 'fps': int_or_none(fmt.get('fps')),
1947 'height': int_or_none(fmt.get('height')),
1948 # bitrate for itag 43 is always 2147483647
1949 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1950 'width': int_or_none(fmt.get('width')),
1951 }
1952
1953 for fmt in streaming_formats:
1954 if fmt.get('drmFamilies') or fmt.get('drm_families'):
1955 continue
1956 url = url_or_none(fmt.get('url'))
1957
1958 if not url:
1959 cipher = fmt.get('cipher') or fmt.get('signatureCipher')
1960 if not cipher:
1961 continue
1962 url_data = compat_parse_qs(cipher)
1963 url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str))
1964 if not url:
1965 continue
1966 else:
1967 cipher = None
1968 url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
1969
1970 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1971 # Unsupported FORMAT_STREAM_TYPE_OTF
1972 if stream_type == 3:
1973 continue
1974
1975 format_id = fmt.get('itag') or url_data['itag'][0]
1976 if not format_id:
1977 continue
1978 format_id = compat_str(format_id)
1979
1980 if cipher:
1981 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1982 ASSETS_RE = (
1983 r'<script[^>]+\bsrc=("[^"]+")[^>]+\bname=["\']player_ias/base',
1984 r'"jsUrl"\s*:\s*("[^"]+")',
1985 r'"assets":.+?"js":\s*("[^"]+")')
1986 jsplayer_url_json = self._search_regex(
1987 ASSETS_RE,
1988 embed_webpage if age_gate else video_webpage,
1989 'JS player URL (1)', default=None)
1990 if not jsplayer_url_json and not age_gate:
1991 # We need the embed website after all
1992 if embed_webpage is None:
1993 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1994 embed_webpage = self._download_webpage(
1995 embed_url, video_id, 'Downloading embed webpage')
1996 jsplayer_url_json = self._search_regex(
1997 ASSETS_RE, embed_webpage, 'JS player URL')
1998
1999 player_url = json.loads(jsplayer_url_json)
2000 if player_url is None:
2001 player_url_json = self._search_regex(
2002 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
2003 video_webpage, 'age gate player URL')
2004 player_url = json.loads(player_url_json)
2005
2006 if 'sig' in url_data:
2007 url += '&signature=' + url_data['sig'][0]
2008 elif 's' in url_data:
2009 encrypted_sig = url_data['s'][0]
2010
2011 if self._downloader.params.get('verbose'):
2012 if player_url is None:
2013 player_desc = 'unknown'
2014 else:
2015 player_type, player_version = self._extract_player_info(player_url)
2016 player_desc = '%s player %s' % ('flash' if player_type == 'swf' else 'html5', player_version)
2017 parts_sizes = self._signature_cache_id(encrypted_sig)
2018 self.to_screen('{%s} signature length %s, %s' %
2019 (format_id, parts_sizes, player_desc))
2020
2021 signature = self._decrypt_signature(
2022 encrypted_sig, video_id, player_url, age_gate)
2023 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2024 url += '&%s=%s' % (sp, signature)
2025 if 'ratebypass' not in url:
2026 url += '&ratebypass=yes'
2027
2028 dct = {
2029 'format_id': format_id,
2030 'url': url,
2031 'player_url': player_url,
2032 }
2033 if format_id in self._formats:
2034 dct.update(self._formats[format_id])
2035 if format_id in formats_spec:
2036 dct.update(formats_spec[format_id])
2037
2038 # Some itags are not included in DASH manifest thus corresponding formats will
2039 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2040 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2041 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2042 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2043
2044 if width is None:
2045 width = int_or_none(fmt.get('width'))
2046 if height is None:
2047 height = int_or_none(fmt.get('height'))
2048
2049 filesize = int_or_none(url_data.get(
2050 'clen', [None])[0]) or _extract_filesize(url)
2051
2052 quality = url_data.get('quality', [None])[0] or fmt.get('quality')
2053 quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel')
2054
2055 tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000)
2056 or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None
2057 fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps'))
2058
2059 more_fields = {
2060 'filesize': filesize,
2061 'tbr': tbr,
2062 'width': width,
2063 'height': height,
2064 'fps': fps,
2065 'format_note': quality_label or quality,
2066 }
2067 for key, value in more_fields.items():
2068 if value:
2069 dct[key] = value
2070 type_ = url_data.get('type', [None])[0] or fmt.get('mimeType')
2071 if type_:
2072 type_split = type_.split(';')
2073 kind_ext = type_split[0].split('/')
2074 if len(kind_ext) == 2:
2075 kind, _ = kind_ext
2076 dct['ext'] = mimetype2ext(type_split[0])
2077 if kind in ('audio', 'video'):
2078 codecs = None
2079 for mobj in re.finditer(
2080 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2081 if mobj.group('key') == 'codecs':
2082 codecs = mobj.group('val')
2083 break
2084 if codecs:
2085 dct.update(parse_codecs(codecs))
2086 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2087 dct['downloader_options'] = {
2088 # Youtube throttles chunks >~10M
2089 'http_chunk_size': 10485760,
2090 }
2091 formats.append(dct)
2092 else:
2093 manifest_url = (
2094 url_or_none(try_get(
2095 player_response,
2096 lambda x: x['streamingData']['hlsManifestUrl'],
2097 compat_str))
2098 or url_or_none(try_get(
2099 video_info, lambda x: x['hlsvp'][0], compat_str)))
2100 if manifest_url:
2101 formats = []
2102 m3u8_formats = self._extract_m3u8_formats(
2103 manifest_url, video_id, 'mp4', fatal=False)
2104 for a_format in m3u8_formats:
2105 itag = self._search_regex(
2106 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2107 if itag:
2108 a_format['format_id'] = itag
2109 if itag in self._formats:
2110 dct = self._formats[itag].copy()
2111 dct.update(a_format)
2112 a_format = dct
2113 a_format['player_url'] = player_url
2114 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2115 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2116 if self._downloader.params.get('youtube_include_hls_manifest', True):
2117 formats.append(a_format)
2118 else:
2119 error_message = extract_unavailable_message()
2120 if not error_message:
2121 reason_list = try_get(
2122 player_response,
2123 lambda x: x['playabilityStatus']['errorScreen']['playerErrorMessageRenderer']['subreason']['runs'],
2124 list) or []
2125 for reason in reason_list:
2126 if not isinstance(reason, dict):
2127 continue
2128 reason_text = try_get(reason, lambda x: x['text'], compat_str)
2129 if reason_text:
2130 if not error_message:
2131 error_message = ''
2132 error_message += reason_text
2133 if error_message:
2134 error_message = clean_html(error_message)
2135 if not error_message:
2136 error_message = clean_html(try_get(
2137 player_response, lambda x: x['playabilityStatus']['reason'],
2138 compat_str))
2139 if not error_message:
2140 error_message = clean_html(
2141 try_get(video_info, lambda x: x['reason'][0], compat_str))
2142 if error_message:
2143 raise ExtractorError(error_message, expected=True)
2144 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2145
2146 # uploader
2147 video_uploader = try_get(
2148 video_info, lambda x: x['author'][0],
2149 compat_str) or str_or_none(video_details.get('author'))
2150 if video_uploader:
2151 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2152 else:
2153 self._downloader.report_warning('unable to extract uploader name')
2154
2155 # uploader_id
2156 video_uploader_id = None
2157 video_uploader_url = None
2158 mobj = re.search(
2159 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2160 video_webpage)
2161 if mobj is not None:
2162 video_uploader_id = mobj.group('uploader_id')
2163 video_uploader_url = mobj.group('uploader_url')
2164 else:
2165 owner_profile_url = url_or_none(microformat.get('ownerProfileUrl'))
2166 if owner_profile_url:
2167 video_uploader_id = self._search_regex(
2168 r'(?:user|channel)/([^/]+)', owner_profile_url, 'uploader id',
2169 default=None)
2170 video_uploader_url = owner_profile_url
2171
2172 channel_id = (
2173 str_or_none(video_details.get('channelId'))
2174 or self._html_search_meta(
2175 'channelId', video_webpage, 'channel id', default=None)
2176 or self._search_regex(
2177 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2178 video_webpage, 'channel id', default=None, group='id'))
2179 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2180
2181 thumbnails = []
2182 thumbnails_list = try_get(
2183 video_details, lambda x: x['thumbnail']['thumbnails'], list) or []
2184 for t in thumbnails_list:
2185 if not isinstance(t, dict):
2186 continue
2187 thumbnail_url = url_or_none(t.get('url'))
2188 if not thumbnail_url:
2189 continue
2190 thumbnails.append({
2191 'url': thumbnail_url,
2192 'width': int_or_none(t.get('width')),
2193 'height': int_or_none(t.get('height')),
2194 })
2195
2196 if not thumbnails:
2197 video_thumbnail = None
2198 # We try first to get a high quality image:
2199 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2200 video_webpage, re.DOTALL)
2201 if m_thumb is not None:
2202 video_thumbnail = m_thumb.group(1)
2203 thumbnail_url = try_get(video_info, lambda x: x['thumbnail_url'][0], compat_str)
2204 if thumbnail_url:
2205 video_thumbnail = compat_urllib_parse_unquote_plus(thumbnail_url)
2206 if video_thumbnail:
2207 thumbnails.append({'url': video_thumbnail})
2208
2209 # upload date
2210 upload_date = self._html_search_meta(
2211 'datePublished', video_webpage, 'upload date', default=None)
2212 if not upload_date:
2213 upload_date = self._search_regex(
2214 [r'(?s)id="eow-date.*?>(.*?)</span>',
2215 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2216 video_webpage, 'upload date', default=None)
2217 if not upload_date:
2218 upload_date = microformat.get('publishDate') or microformat.get('uploadDate')
2219 upload_date = unified_strdate(upload_date)
2220
2221 video_license = self._html_search_regex(
2222 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2223 video_webpage, 'license', default=None)
2224
2225 m_music = re.search(
2226 r'''(?x)
2227 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2228 <ul[^>]*>\s*
2229 <li>(?P<title>.+?)
2230 by (?P<creator>.+?)
2231 (?:
2232 \(.+?\)|
2233 <a[^>]*
2234 (?:
2235 \bhref=["\']/red[^>]*>| # drop possible
2236 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2237 )
2238 .*?
2239 )?</li
2240 ''',
2241 video_webpage)
2242 if m_music:
2243 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2244 video_creator = clean_html(m_music.group('creator'))
2245 else:
2246 video_alt_title = video_creator = None
2247
2248 def extract_meta(field):
2249 return self._html_search_regex(
2250 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2251 video_webpage, field, default=None)
2252
2253 track = extract_meta('Song')
2254 artist = extract_meta('Artist')
2255 album = extract_meta('Album')
2256
2257 # Youtube Music Auto-generated description
2258 release_date = release_year = None
2259 if video_description:
2260 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)
2261 if mobj:
2262 if not track:
2263 track = mobj.group('track').strip()
2264 if not artist:
2265 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2266 if not album:
2267 album = mobj.group('album'.strip())
2268 release_year = mobj.group('release_year')
2269 release_date = mobj.group('release_date')
2270 if release_date:
2271 release_date = release_date.replace('-', '')
2272 if not release_year:
2273 release_year = int(release_date[:4])
2274 if release_year:
2275 release_year = int(release_year)
2276
2277 yt_initial_data = self._extract_yt_initial_data(video_id, video_webpage)
2278 contents = try_get(yt_initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
2279 for content in contents:
2280 rows = try_get(content, lambda x: x['videoSecondaryInfoRenderer']['metadataRowContainer']['metadataRowContainerRenderer']['rows'], list) or []
2281 multiple_songs = False
2282 for row in rows:
2283 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2284 multiple_songs = True
2285 break
2286 for row in rows:
2287 mrr = row.get('metadataRowRenderer') or {}
2288 mrr_title = try_get(
2289 mrr, lambda x: x['title']['simpleText'], compat_str)
2290 mrr_contents = try_get(
2291 mrr, lambda x: x['contents'][0], dict) or {}
2292 mrr_contents_text = try_get(mrr_contents, [lambda x: x['simpleText'], lambda x: x['runs'][0]['text']], compat_str)
2293 if not (mrr_title and mrr_contents_text):
2294 continue
2295 if mrr_title == 'License':
2296 video_license = mrr_contents_text
2297 elif not multiple_songs:
2298 if mrr_title == 'Album':
2299 album = mrr_contents_text
2300 elif mrr_title == 'Artist':
2301 artist = mrr_contents_text
2302 elif mrr_title == 'Song':
2303 track = mrr_contents_text
2304
2305 m_episode = re.search(
2306 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2307 video_webpage)
2308 if m_episode:
2309 series = unescapeHTML(m_episode.group('series'))
2310 season_number = int(m_episode.group('season'))
2311 episode_number = int(m_episode.group('episode'))
2312 else:
2313 series = season_number = episode_number = None
2314
2315 m_cat_container = self._search_regex(
2316 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2317 video_webpage, 'categories', default=None)
2318 category = None
2319 if m_cat_container:
2320 category = self._html_search_regex(
2321 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2322 default=None)
2323 if not category:
2324 category = try_get(
2325 microformat, lambda x: x['category'], compat_str)
2326 video_categories = None if category is None else [category]
2327
2328 video_tags = [
2329 unescapeHTML(m.group('content'))
2330 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2331 if not video_tags:
2332 video_tags = try_get(video_details, lambda x: x['keywords'], list)
2333
2334 def _extract_count(count_name):
2335 return str_to_int(self._search_regex(
2336 (r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>' % re.escape(count_name),
2337 r'["\']label["\']\s*:\s*["\']([\d,.]+)\s+%ss["\']' % re.escape(count_name)),
2338 video_webpage, count_name, default=None))
2339
2340 like_count = _extract_count('like')
2341 dislike_count = _extract_count('dislike')
2342
2343 if view_count is None:
2344 view_count = str_to_int(self._search_regex(
2345 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2346 'view count', default=None))
2347
2348 average_rating = (
2349 float_or_none(video_details.get('averageRating'))
2350 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2351
2352 # subtitles
2353 video_subtitles = self.extract_subtitles(
2354 video_id, video_webpage, has_live_chat_replay)
2355 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2356
2357 video_duration = try_get(
2358 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2359 if not video_duration:
2360 video_duration = int_or_none(video_details.get('lengthSeconds'))
2361 if not video_duration:
2362 video_duration = parse_duration(self._html_search_meta(
2363 'duration', video_webpage, 'video duration'))
2364
2365 # Get Subscriber Count of channel
2366 subscriber_count = parse_count(self._search_regex(
2367 r'"text":"([\d\.]+\w?) subscribers"',
2368 video_webpage,
2369 'subscriber count',
2370 default=None
2371 ))
2372
2373 # annotations
2374 video_annotations = None
2375 if self._downloader.params.get('writeannotations', False):
2376 xsrf_token = self._search_regex(
2377 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2378 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2379 invideo_url = try_get(
2380 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2381 if xsrf_token and invideo_url:
2382 xsrf_field_name = self._search_regex(
2383 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2384 video_webpage, 'xsrf field name',
2385 group='xsrf_field_name', default='session_token')
2386 video_annotations = self._download_webpage(
2387 self._proto_relative_url(invideo_url),
2388 video_id, note='Downloading annotations',
2389 errnote='Unable to download video annotations', fatal=False,
2390 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2391
2392 chapters = self._extract_chapters(video_webpage, description_original, video_id, video_duration)
2393
2394 # Look for the DASH manifest
2395 if self._downloader.params.get('youtube_include_dash_manifest', True):
2396 dash_mpd_fatal = True
2397 for mpd_url in dash_mpds:
2398 dash_formats = {}
2399 try:
2400 def decrypt_sig(mobj):
2401 s = mobj.group(1)
2402 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2403 return '/signature/%s' % dec_s
2404
2405 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2406
2407 for df in self._extract_mpd_formats(
2408 mpd_url, video_id, fatal=dash_mpd_fatal,
2409 formats_dict=self._formats):
2410 if not df.get('filesize'):
2411 df['filesize'] = _extract_filesize(df['url'])
2412 # Do not overwrite DASH format found in some previous DASH manifest
2413 if df['format_id'] not in dash_formats:
2414 dash_formats[df['format_id']] = df
2415 # Additional DASH manifests may end up in HTTP Error 403 therefore
2416 # allow them to fail without bug report message if we already have
2417 # some DASH manifest succeeded. This is temporary workaround to reduce
2418 # burst of bug reports until we figure out the reason and whether it
2419 # can be fixed at all.
2420 dash_mpd_fatal = False
2421 except (ExtractorError, KeyError) as e:
2422 self.report_warning(
2423 'Skipping DASH manifest: %r' % e, video_id)
2424 if dash_formats:
2425 # Remove the formats we found through non-DASH, they
2426 # contain less info and it can be wrong, because we use
2427 # fixed values (for example the resolution). See
2428 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2429 # example.
2430 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2431 formats.extend(dash_formats.values())
2432
2433 # Check for malformed aspect ratio
2434 stretched_m = re.search(
2435 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2436 video_webpage)
2437 if stretched_m:
2438 w = float(stretched_m.group('w'))
2439 h = float(stretched_m.group('h'))
2440 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2441 # We will only process correct ratios.
2442 if w > 0 and h > 0:
2443 ratio = w / h
2444 for f in formats:
2445 if f.get('vcodec') != 'none':
2446 f['stretched_ratio'] = ratio
2447
2448 if not formats:
2449 if 'reason' in video_info:
2450 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2451 regions_allowed = self._html_search_meta(
2452 'regionsAllowed', video_webpage, default=None)
2453 countries = regions_allowed.split(',') if regions_allowed else None
2454 self.raise_geo_restricted(
2455 msg=video_info['reason'][0], countries=countries)
2456 reason = video_info['reason'][0]
2457 if 'Invalid parameters' in reason:
2458 unavailable_message = extract_unavailable_message()
2459 if unavailable_message:
2460 reason = unavailable_message
2461 raise ExtractorError(
2462 'YouTube said: %s' % reason,
2463 expected=True, video_id=video_id)
2464 if video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos']):
2465 raise ExtractorError('This video is DRM protected.', expected=True)
2466
2467 self._sort_formats(formats)
2468
2469 self.mark_watched(video_id, video_info, player_response)
2470
2471 return {
2472 'id': video_id,
2473 'uploader': video_uploader,
2474 'uploader_id': video_uploader_id,
2475 'uploader_url': video_uploader_url,
2476 'channel_id': channel_id,
2477 'channel_url': channel_url,
2478 'upload_date': upload_date,
2479 'license': video_license,
2480 'creator': video_creator or artist,
2481 'title': video_title,
2482 'alt_title': video_alt_title or track,
2483 'thumbnails': thumbnails,
2484 'description': video_description,
2485 'categories': video_categories,
2486 'tags': video_tags,
2487 'subtitles': video_subtitles,
2488 'automatic_captions': automatic_captions,
2489 'duration': video_duration,
2490 'age_limit': 18 if age_gate else 0,
2491 'annotations': video_annotations,
2492 'chapters': chapters,
2493 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2494 'view_count': view_count,
2495 'like_count': like_count,
2496 'dislike_count': dislike_count,
2497 'average_rating': average_rating,
2498 'formats': formats,
2499 'is_live': is_live,
2500 'start_time': start_time,
2501 'end_time': end_time,
2502 'series': series,
2503 'season_number': season_number,
2504 'episode_number': episode_number,
2505 'track': track,
2506 'artist': artist,
2507 'album': album,
2508 'release_date': release_date,
2509 'release_year': release_year,
2510 'subscriber_count': subscriber_count,
2511 }
2512
2513
2514 class YoutubeTabIE(YoutubeBaseInfoExtractor):
2515 IE_DESC = 'YouTube.com tab'
2516 _VALID_URL = r'''(?x)
2517 https?://
2518 (?:\w+\.)?
2519 (?:
2520 youtube(?:kids)?\.com|
2521 invidio\.us
2522 )/
2523 (?:
2524 (?:channel|c|user)/|
2525 (?P<not_channel>
2526 feed/|
2527 (?:playlist|watch)\?.*?\blist=
2528 )|
2529 (?!(%s)([/#?]|$)) # Direct URLs
2530 )
2531 (?P<id>[^/?\#&]+)
2532 ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
2533 IE_NAME = 'youtube:tab'
2534
2535 _TESTS = [{
2536 # playlists, multipage
2537 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
2538 'playlist_mincount': 94,
2539 'info_dict': {
2540 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2541 'title': 'Игорь Клейнер - Playlists',
2542 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2543 },
2544 }, {
2545 # playlists, multipage, different order
2546 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2547 'playlist_mincount': 94,
2548 'info_dict': {
2549 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2550 'title': 'Игорь Клейнер - Playlists',
2551 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2552 },
2553 }, {
2554 # playlists, singlepage
2555 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2556 'playlist_mincount': 4,
2557 'info_dict': {
2558 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2559 'title': 'ThirstForScience - Playlists',
2560 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
2561 }
2562 }, {
2563 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
2564 'only_matching': True,
2565 }, {
2566 # basic, single video playlist
2567 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2568 'info_dict': {
2569 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2570 'uploader': 'Sergey M.',
2571 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2572 'title': 'youtube-dl public playlist',
2573 },
2574 'playlist_count': 1,
2575 }, {
2576 # empty playlist
2577 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2578 'info_dict': {
2579 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2580 'uploader': 'Sergey M.',
2581 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2582 'title': 'youtube-dl empty playlist',
2583 },
2584 'playlist_count': 0,
2585 }, {
2586 # Home tab
2587 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
2588 'info_dict': {
2589 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2590 'title': 'lex will - Home',
2591 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2592 },
2593 'playlist_mincount': 2,
2594 }, {
2595 # Videos tab
2596 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
2597 'info_dict': {
2598 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2599 'title': 'lex will - Videos',
2600 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2601 },
2602 'playlist_mincount': 975,
2603 }, {
2604 # Videos tab, sorted by popular
2605 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
2606 'info_dict': {
2607 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2608 'title': 'lex will - Videos',
2609 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2610 },
2611 'playlist_mincount': 199,
2612 }, {
2613 # Playlists tab
2614 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
2615 'info_dict': {
2616 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2617 'title': 'lex will - Playlists',
2618 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2619 },
2620 'playlist_mincount': 17,
2621 }, {
2622 # Community tab
2623 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
2624 'info_dict': {
2625 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2626 'title': 'lex will - Community',
2627 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2628 },
2629 'playlist_mincount': 18,
2630 }, {
2631 # Channels tab
2632 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
2633 'info_dict': {
2634 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2635 'title': 'lex will - Channels',
2636 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2637 },
2638 'playlist_mincount': 138,
2639 }, {
2640 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2641 'only_matching': True,
2642 }, {
2643 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2644 'only_matching': True,
2645 }, {
2646 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2647 'only_matching': True,
2648 }, {
2649 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2650 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2651 'info_dict': {
2652 'title': '29C3: Not my department',
2653 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2654 'uploader': 'Christiaan008',
2655 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
2656 },
2657 'playlist_count': 96,
2658 }, {
2659 'note': 'Large playlist',
2660 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2661 'info_dict': {
2662 'title': 'Uploads from Cauchemar',
2663 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2664 'uploader': 'Cauchemar',
2665 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
2666 },
2667 'playlist_mincount': 1123,
2668 }, {
2669 # even larger playlist, 8832 videos
2670 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
2671 'only_matching': True,
2672 }, {
2673 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2674 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2675 'info_dict': {
2676 'title': 'Uploads from Interstellar Movie',
2677 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2678 'uploader': 'Interstellar Movie',
2679 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
2680 },
2681 'playlist_mincount': 21,
2682 }, {
2683 # https://github.com/ytdl-org/youtube-dl/issues/21844
2684 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2685 'info_dict': {
2686 'title': 'Data Analysis with Dr Mike Pound',
2687 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2688 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
2689 'uploader': 'Computerphile',
2690 },
2691 'playlist_mincount': 11,
2692 }, {
2693 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2694 'only_matching': True,
2695 }, {
2696 # Playlist URL that does not actually serve a playlist
2697 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2698 'info_dict': {
2699 'id': 'FqZTN594JQw',
2700 'ext': 'webm',
2701 'title': "Smiley's People 01 detective, Adventure Series, Action",
2702 'uploader': 'STREEM',
2703 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2704 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2705 'upload_date': '20150526',
2706 'license': 'Standard YouTube License',
2707 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2708 'categories': ['People & Blogs'],
2709 'tags': list,
2710 'view_count': int,
2711 'like_count': int,
2712 'dislike_count': int,
2713 },
2714 'params': {
2715 'skip_download': True,
2716 },
2717 'skip': 'This video is not available.',
2718 'add_ie': [YoutubeIE.ie_key()],
2719 }, {
2720 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2721 'only_matching': True,
2722 }, {
2723 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
2724 'only_matching': True,
2725 }, {
2726 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
2727 'info_dict': {
2728 'id': '9Auq9mYxFEE',
2729 'ext': 'mp4',
2730 'title': 'Watch Sky News live',
2731 'uploader': 'Sky News',
2732 'uploader_id': 'skynews',
2733 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
2734 'upload_date': '20191102',
2735 'description': 'md5:78de4e1c2359d0ea3ed829678e38b662',
2736 'categories': ['News & Politics'],
2737 'tags': list,
2738 'like_count': int,
2739 'dislike_count': int,
2740 },
2741 'params': {
2742 'skip_download': True,
2743 },
2744 }, {
2745 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2746 'info_dict': {
2747 'id': 'a48o2S1cPoo',
2748 'ext': 'mp4',
2749 'title': 'The Young Turks - Live Main Show',
2750 'uploader': 'The Young Turks',
2751 'uploader_id': 'TheYoungTurks',
2752 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2753 'upload_date': '20150715',
2754 'license': 'Standard YouTube License',
2755 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2756 'categories': ['News & Politics'],
2757 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2758 'like_count': int,
2759 'dislike_count': int,
2760 },
2761 'params': {
2762 'skip_download': True,
2763 },
2764 'only_matching': True,
2765 }, {
2766 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2767 'only_matching': True,
2768 }, {
2769 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2770 'only_matching': True,
2771 }, {
2772 'url': 'https://www.youtube.com/feed/trending',
2773 'only_matching': True,
2774 }, {
2775 # needs auth
2776 'url': 'https://www.youtube.com/feed/library',
2777 'only_matching': True,
2778 }, {
2779 # needs auth
2780 'url': 'https://www.youtube.com/feed/history',
2781 'only_matching': True,
2782 }, {
2783 # needs auth
2784 'url': 'https://www.youtube.com/feed/subscriptions',
2785 'only_matching': True,
2786 }, {
2787 # needs auth
2788 'url': 'https://www.youtube.com/feed/watch_later',
2789 'only_matching': True,
2790 }, {
2791 # no longer available?
2792 'url': 'https://www.youtube.com/feed/recommended',
2793 'only_matching': True,
2794 }
2795 # TODO
2796 # {
2797 # 'url': 'https://www.youtube.com/TheYoungTurks/live',
2798 # 'only_matching': True,
2799 # }
2800 ]
2801
2802 def _extract_channel_id(self, webpage):
2803 channel_id = self._html_search_meta(
2804 'channelId', webpage, 'channel id', default=None)
2805 if channel_id:
2806 return channel_id
2807 channel_url = self._html_search_meta(
2808 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
2809 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
2810 'twitter:app:url:googleplay'), webpage, 'channel url')
2811 return self._search_regex(
2812 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
2813 channel_url, 'channel id')
2814
2815 @staticmethod
2816 def _extract_grid_item_renderer(item):
2817 for item_kind in ('Playlist', 'Video', 'Channel'):
2818 renderer = item.get('grid%sRenderer' % item_kind)
2819 if renderer:
2820 return renderer
2821
2822 def _extract_video(self, renderer):
2823 video_id = renderer.get('videoId')
2824 title = try_get(
2825 renderer,
2826 (lambda x: x['title']['runs'][0]['text'],
2827 lambda x: x['title']['simpleText']), compat_str)
2828 description = try_get(
2829 renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'],
2830 compat_str)
2831 duration = parse_duration(try_get(
2832 renderer, lambda x: x['lengthText']['simpleText'], compat_str))
2833 view_count_text = try_get(
2834 renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
2835 view_count = str_to_int(self._search_regex(
2836 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
2837 'view count', default=None))
2838 uploader = try_get(
2839 renderer, lambda x: x['ownerText']['runs'][0]['text'], compat_str)
2840 return {
2841 '_type': 'url_transparent',
2842 'ie_key': YoutubeIE.ie_key(),
2843 'id': video_id,
2844 'url': video_id,
2845 'title': title,
2846 'description': description,
2847 'duration': duration,
2848 'view_count': view_count,
2849 'uploader': uploader,
2850 }
2851
2852 def _grid_entries(self, grid_renderer):
2853 for item in grid_renderer['items']:
2854 if not isinstance(item, dict):
2855 continue
2856 renderer = self._extract_grid_item_renderer(item)
2857 if not isinstance(renderer, dict):
2858 continue
2859 title = try_get(
2860 renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2861 # playlist
2862 playlist_id = renderer.get('playlistId')
2863 if playlist_id:
2864 yield self.url_result(
2865 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2866 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
2867 video_title=title)
2868 # video
2869 video_id = renderer.get('videoId')
2870 if video_id:
2871 yield self._extract_video(renderer)
2872 # channel
2873 channel_id = renderer.get('channelId')
2874 if channel_id:
2875 title = try_get(
2876 renderer, lambda x: x['title']['simpleText'], compat_str)
2877 yield self.url_result(
2878 'https://www.youtube.com/channel/%s' % channel_id,
2879 ie=YoutubeTabIE.ie_key(), video_title=title)
2880
2881 def _shelf_entries_from_content(self, shelf_renderer):
2882 content = shelf_renderer.get('content')
2883 if not isinstance(content, dict):
2884 return
2885 renderer = content.get('gridRenderer')
2886 if renderer:
2887 # TODO: add support for nested playlists so each shelf is processed
2888 # as separate playlist
2889 # TODO: this includes only first N items
2890 for entry in self._grid_entries(renderer):
2891 yield entry
2892 renderer = content.get('horizontalListRenderer')
2893 if renderer:
2894 # TODO
2895 pass
2896
2897 def _shelf_entries(self, shelf_renderer):
2898 ep = try_get(
2899 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
2900 compat_str)
2901 shelf_url = urljoin('https://www.youtube.com', ep)
2902 if shelf_url:
2903 title = try_get(
2904 shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2905 yield self.url_result(shelf_url, video_title=title)
2906 # Shelf may not contain shelf URL, fallback to extraction from content
2907 for entry in self._shelf_entries_from_content(shelf_renderer):
2908 yield entry
2909
2910 def _playlist_entries(self, video_list_renderer):
2911 for content in video_list_renderer['contents']:
2912 if not isinstance(content, dict):
2913 continue
2914 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
2915 if not isinstance(renderer, dict):
2916 continue
2917 video_id = renderer.get('videoId')
2918 if not video_id:
2919 continue
2920 yield self._extract_video(renderer)
2921
2922 r""" # Not needed in the new implementation
2923 def _itemSection_entries(self, item_sect_renderer):
2924 for content in item_sect_renderer['contents']:
2925 if not isinstance(content, dict):
2926 continue
2927 renderer = content.get('videoRenderer', {})
2928 if not isinstance(renderer, dict):
2929 continue
2930 video_id = renderer.get('videoId')
2931 if not video_id:
2932 continue
2933 yield self._extract_video(renderer)
2934 """
2935
2936 def _rich_entries(self, rich_grid_renderer):
2937 renderer = try_get(
2938 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
2939 video_id = renderer.get('videoId')
2940 if not video_id:
2941 return
2942 yield self._extract_video(renderer)
2943
2944 def _video_entry(self, video_renderer):
2945 video_id = video_renderer.get('videoId')
2946 if video_id:
2947 return self._extract_video(video_renderer)
2948
2949 def _post_thread_entries(self, post_thread_renderer):
2950 post_renderer = try_get(
2951 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
2952 if not post_renderer:
2953 return
2954 # video attachment
2955 video_renderer = try_get(
2956 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict)
2957 video_id = None
2958 if video_renderer:
2959 entry = self._video_entry(video_renderer)
2960 if entry:
2961 yield entry
2962 # inline video links
2963 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
2964 for run in runs:
2965 if not isinstance(run, dict):
2966 continue
2967 ep_url = try_get(
2968 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
2969 if not ep_url:
2970 continue
2971 if not YoutubeIE.suitable(ep_url):
2972 continue
2973 ep_video_id = YoutubeIE._match_id(ep_url)
2974 if video_id == ep_video_id:
2975 continue
2976 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=video_id)
2977
2978 def _post_thread_continuation_entries(self, post_thread_continuation):
2979 contents = post_thread_continuation.get('contents')
2980 if not isinstance(contents, list):
2981 return
2982 for content in contents:
2983 renderer = content.get('backstagePostThreadRenderer')
2984 if not isinstance(renderer, dict):
2985 continue
2986 for entry in self._post_thread_entries(renderer):
2987 yield entry
2988
2989 @staticmethod
2990 def _extract_next_continuation_data(renderer):
2991 next_continuation = try_get(
2992 renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict)
2993 if not next_continuation:
2994 return
2995 continuation = next_continuation.get('continuation')
2996 if not continuation:
2997 return
2998 ctp = next_continuation.get('clickTrackingParams')
2999 return {
3000 'ctoken': continuation,
3001 'continuation': continuation,
3002 'itct': ctp,
3003 }
3004
3005 @classmethod
3006 def _extract_continuation(cls, renderer):
3007 next_continuation = cls._extract_next_continuation_data(renderer)
3008 if next_continuation:
3009 return next_continuation
3010 contents = renderer.get('contents')
3011 if not isinstance(contents, list):
3012 return
3013 for content in contents:
3014 if not isinstance(content, dict):
3015 continue
3016 continuation_ep = try_get(
3017 content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
3018 dict)
3019 if not continuation_ep:
3020 continue
3021 continuation = try_get(
3022 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
3023 if not continuation:
3024 continue
3025 ctp = continuation_ep.get('clickTrackingParams')
3026 if not ctp:
3027 continue
3028 return {
3029 'ctoken': continuation,
3030 'continuation': continuation,
3031 'itct': ctp,
3032 }
3033
3034 def _entries(self, tab, identity_token):
3035
3036 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
3037 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
3038 for content in contents:
3039 if not isinstance(content, dict):
3040 continue
3041 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
3042 if not is_renderer:
3043 renderer = content.get('richItemRenderer')
3044 if renderer:
3045 for entry in self._rich_entries(renderer):
3046 yield entry
3047 continuation_list[0] = self._extract_continuation(parent_renderer)
3048 continue
3049 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
3050 for isr_content in isr_contents:
3051 if not isinstance(isr_content, dict):
3052 continue
3053 renderer = isr_content.get('playlistVideoListRenderer')
3054 if renderer:
3055 for entry in self._playlist_entries(renderer):
3056 yield entry
3057 continuation_list[0] = self._extract_continuation(renderer)
3058 continue
3059 renderer = isr_content.get('gridRenderer')
3060 if renderer:
3061 for entry in self._grid_entries(renderer):
3062 yield entry
3063 continuation_list[0] = self._extract_continuation(renderer)
3064 continue
3065 renderer = isr_content.get('shelfRenderer')
3066 if renderer:
3067 for entry in self._shelf_entries(renderer):
3068 yield entry
3069 continue
3070 renderer = isr_content.get('backstagePostThreadRenderer')
3071 if renderer:
3072 for entry in self._post_thread_entries(renderer):
3073 yield entry
3074 continuation_list[0] = self._extract_continuation(renderer)
3075 continue
3076 renderer = isr_content.get('videoRenderer')
3077 if renderer:
3078 entry = self._video_entry(renderer)
3079 if entry:
3080 yield entry
3081
3082 if not continuation_list[0]:
3083 continuation_list[0] = self._extract_continuation(is_renderer)
3084
3085 if not continuation_list[0]:
3086 continuation_list[0] = self._extract_continuation(parent_renderer)
3087
3088 continuation_list = [None] # Python 2 doesnot support nonlocal
3089 parent_renderer = (
3090 try_get(tab, lambda x: x['sectionListRenderer'], dict)
3091 or try_get(tab, lambda x: x['richGridRenderer'], dict) or {})
3092 for entry in extract_entries(parent_renderer):
3093 yield entry
3094 continuation = continuation_list[0]
3095
3096 headers = {
3097 'x-youtube-client-name': '1',
3098 'x-youtube-client-version': '2.20201112.04.01',
3099 }
3100 if identity_token:
3101 headers['x-youtube-identity-token'] = identity_token
3102
3103 for page_num in itertools.count(1):
3104 if not continuation:
3105 break
3106 browse = self._download_json(
3107 'https://www.youtube.com/browse_ajax', None,
3108 'Downloading page %d' % page_num,
3109 headers=headers, query=continuation, fatal=False)
3110 if not browse:
3111 break
3112 response = try_get(browse, lambda x: x[1]['response'], dict)
3113 if not response:
3114 break
3115
3116 continuation_contents = try_get(
3117 response, lambda x: x['continuationContents'], dict)
3118 if continuation_contents:
3119 continuation_renderer = continuation_contents.get('playlistVideoListContinuation')
3120 if continuation_renderer:
3121 for entry in self._playlist_entries(continuation_renderer):
3122 yield entry
3123 continuation = self._extract_continuation(continuation_renderer)
3124 continue
3125 continuation_renderer = continuation_contents.get('gridContinuation')
3126 if continuation_renderer:
3127 for entry in self._grid_entries(continuation_renderer):
3128 yield entry
3129 continuation = self._extract_continuation(continuation_renderer)
3130 continue
3131 continuation_renderer = continuation_contents.get('itemSectionContinuation')
3132 if continuation_renderer:
3133 for entry in self._post_thread_continuation_entries(continuation_renderer):
3134 yield entry
3135 continuation = self._extract_continuation(continuation_renderer)
3136 continue
3137 continuation_renderer = continuation_contents.get('sectionListContinuation') # for feeds
3138 if continuation_renderer:
3139 continuation_list = [None]
3140 for entry in extract_entries(continuation_renderer):
3141 yield entry
3142 continuation = continuation_list[0]
3143 continue
3144
3145 continuation_items = try_get(
3146 response, lambda x: x['onResponseReceivedActions'][0]['appendContinuationItemsAction']['continuationItems'], list)
3147 if continuation_items:
3148 continuation_item = continuation_items[0]
3149 if not isinstance(continuation_item, dict):
3150 continue
3151 renderer = continuation_item.get('playlistVideoRenderer') or continuation_item.get('itemSectionRenderer')
3152 if renderer:
3153 video_list_renderer = {'contents': continuation_items}
3154 for entry in self._playlist_entries(video_list_renderer):
3155 yield entry
3156 continuation = self._extract_continuation(video_list_renderer)
3157 continue
3158 break
3159
3160 @staticmethod
3161 def _extract_selected_tab(tabs):
3162 for tab in tabs:
3163 if try_get(tab, lambda x: x['tabRenderer']['selected'], bool):
3164 return tab['tabRenderer']
3165 else:
3166 raise ExtractorError('Unable to find selected tab')
3167
3168 @staticmethod
3169 def _extract_uploader(data):
3170 uploader = {}
3171 sidebar_renderer = try_get(
3172 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3173 if sidebar_renderer:
3174 for item in sidebar_renderer:
3175 if not isinstance(item, dict):
3176 continue
3177 renderer = item.get('playlistSidebarSecondaryInfoRenderer')
3178 if not isinstance(renderer, dict):
3179 continue
3180 owner = try_get(
3181 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3182 if owner:
3183 uploader['uploader'] = owner.get('text')
3184 uploader['uploader_id'] = try_get(
3185 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3186 uploader['uploader_url'] = urljoin(
3187 'https://www.youtube.com/',
3188 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3189 return uploader
3190
3191 def _extract_from_tabs(self, item_id, webpage, data, tabs, identity_token):
3192 selected_tab = self._extract_selected_tab(tabs)
3193 renderer = try_get(
3194 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3195 playlist_id = title = description = None
3196 if renderer:
3197 channel_title = renderer.get('title') or item_id
3198 tab_title = selected_tab.get('title')
3199 title = channel_title or item_id
3200 if tab_title:
3201 title += ' - %s' % tab_title
3202 description = renderer.get('description')
3203 playlist_id = renderer.get('externalId')
3204 renderer = try_get(
3205 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3206 if renderer:
3207 title = renderer.get('title')
3208 description = None
3209 playlist_id = item_id
3210 if playlist_id is None:
3211 playlist_id = item_id
3212 if title is None:
3213 title = "Youtube " + playlist_id.title()
3214 playlist = self.playlist_result(
3215 self._entries(selected_tab['content'], identity_token),
3216 playlist_id=playlist_id, playlist_title=title,
3217 playlist_description=description)
3218 playlist.update(self._extract_uploader(data))
3219 return playlist
3220
3221 def _extract_from_playlist(self, item_id, data, playlist):
3222 title = playlist.get('title') or try_get(
3223 data, lambda x: x['titleText']['simpleText'], compat_str)
3224 playlist_id = playlist.get('playlistId') or item_id
3225 return self.playlist_result(
3226 self._playlist_entries(playlist), playlist_id=playlist_id,
3227 playlist_title=title)
3228
3229 def _extract_alerts(self, data):
3230 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
3231 for renderer in alert_dict:
3232 alert = alert_dict[renderer]
3233 alert_type = alert.get('type')
3234 if not alert_type:
3235 continue
3236 message = try_get(alert, lambda x: x['text']['simpleText'], compat_str)
3237 if message:
3238 yield alert_type, message
3239 for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
3240 message = try_get(run, lambda x: x['text'], compat_str)
3241 if message:
3242 yield alert_type, message
3243
3244 def _real_extract(self, url):
3245 item_id = self._match_id(url)
3246 url = compat_urlparse.urlunparse(
3247 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
3248 is_home = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
3249 if is_home is not None and is_home.group('not_channel') is None and item_id != 'feed':
3250 self._downloader.report_warning(
3251 'A channel/user page was given. All the channel\'s videos will be downloaded. '
3252 'To download only the videos in the home page, add a "/home" to the URL')
3253 url = '%s/videos%s' % (is_home.group('pre'), is_home.group('post') or '')
3254
3255 # Handle both video/playlist URLs
3256 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3257 video_id = qs.get('v', [None])[0]
3258 playlist_id = qs.get('list', [None])[0]
3259
3260 if is_home.group('not_channel') is not None and is_home.group('not_channel').startswith('watch') and not video_id:
3261 if playlist_id:
3262 self._downloader.report_warning('%s is not a valid Youtube URL. Trying to download playlist %s' % (url, playlist_id))
3263 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
3264 # return self.url_result(playlist_id, ie=YoutubePlaylistIE.ie_key())
3265 else:
3266 raise ExtractorError('Unable to recognize tab page')
3267 if video_id and playlist_id:
3268 if self._downloader.params.get('noplaylist'):
3269 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3270 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3271 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
3272
3273 webpage = self._download_webpage(url, item_id)
3274 identity_token = self._search_regex(
3275 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
3276 'identity token', default=None)
3277 data = self._extract_yt_initial_data(item_id, webpage)
3278 for alert_type, alert_message in self._extract_alerts(data):
3279 self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
3280 tabs = try_get(
3281 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
3282 if tabs:
3283 return self._extract_from_tabs(item_id, webpage, data, tabs, identity_token)
3284 playlist = try_get(
3285 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3286 if playlist:
3287 return self._extract_from_playlist(item_id, data, playlist)
3288 # Fallback to video extraction if no playlist alike page is recognized.
3289 # First check for the current video then try the v attribute of URL query.
3290 video_id = try_get(
3291 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
3292 compat_str) or video_id
3293 if video_id:
3294 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3295 # Failed to recognize
3296 raise ExtractorError('Unable to recognize tab page')
3297
3298
3299 class YoutubePlaylistIE(InfoExtractor):
3300 IE_DESC = 'YouTube.com playlists'
3301 _VALID_URL = r'''(?x)(?:
3302 (?:https?://)?
3303 (?:\w+\.)?
3304 (?:
3305 (?:
3306 youtube(?:kids)?\.com|
3307 invidio\.us|
3308 youtu\.be
3309 )
3310 /.*?\?.*?\blist=
3311 )?
3312 (?P<id>%(playlist_id)s)
3313 )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3314 IE_NAME = 'youtube:playlist'
3315 _TESTS = [{
3316 'note': 'issue #673',
3317 'url': 'PLBB231211A4F62143',
3318 'info_dict': {
3319 'title': '[OLD]Team Fortress 2 (Class-based LP)',
3320 'id': 'PLBB231211A4F62143',
3321 'uploader': 'Wickydoo',
3322 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
3323 },
3324 'playlist_mincount': 29,
3325 }, {
3326 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3327 'info_dict': {
3328 'title': 'YDL_safe_search',
3329 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3330 },
3331 'playlist_count': 2,
3332 'skip': 'This playlist is private',
3333 }, {
3334 'note': 'embedded',
3335 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3336 'playlist_count': 4,
3337 'info_dict': {
3338 'title': 'JODA15',
3339 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3340 'uploader': 'milan',
3341 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
3342 }
3343 }, {
3344 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3345 'playlist_mincount': 982,
3346 'info_dict': {
3347 'title': '2018 Chinese New Singles (11/6 updated)',
3348 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3349 'uploader': 'LBK',
3350 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
3351 }
3352 }, {
3353 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
3354 'info_dict': {
3355 'id': 'yeWKywCrFtk',
3356 'ext': 'mp4',
3357 'title': 'Small Scale Baler and Braiding Rugs',
3358 'uploader': 'Backus-Page House Museum',
3359 'uploader_id': 'backuspagemuseum',
3360 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
3361 'upload_date': '20161008',
3362 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
3363 'categories': ['Nonprofits & Activism'],
3364 'tags': list,
3365 'like_count': int,
3366 'dislike_count': int,
3367 },
3368 'params': {
3369 'noplaylist': True,
3370 'skip_download': True,
3371 },
3372 }, {
3373 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
3374 'only_matching': True,
3375 }, {
3376 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
3377 'only_matching': True,
3378 }, {
3379 # music album playlist
3380 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
3381 'only_matching': True,
3382 }]
3383
3384 @classmethod
3385 def suitable(cls, url):
3386 return False if YoutubeTabIE.suitable(url) else super(
3387 YoutubePlaylistIE, cls).suitable(url)
3388
3389 def _real_extract(self, url):
3390 playlist_id = self._match_id(url)
3391 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3392 if not qs:
3393 qs = {'list': playlist_id}
3394 return self.url_result(
3395 update_url_query('https://www.youtube.com/playlist', qs),
3396 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3397
3398
3399 class YoutubeYtUserIE(InfoExtractor):
3400 _VALID_URL = r'ytuser:(?P<id>.+)'
3401 _TESTS = [{
3402 'url': 'ytuser:phihag',
3403 'only_matching': True,
3404 }]
3405
3406 def _real_extract(self, url):
3407 user_id = self._match_id(url)
3408 return self.url_result(
3409 'https://www.youtube.com/user/%s' % user_id,
3410 ie=YoutubeTabIE.ie_key(), video_id=user_id)
3411
3412
3413 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3414 IE_NAME = 'youtube:favorites'
3415 IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)'
3416 _VALID_URL = r':ytfav(?:ou?rite)?s?'
3417 _LOGIN_REQUIRED = True
3418 _TESTS = [{
3419 'url': ':ytfav',
3420 'only_matching': True,
3421 }, {
3422 'url': ':ytfavorites',
3423 'only_matching': True,
3424 }]
3425
3426 def _real_extract(self, url):
3427 return self.url_result(
3428 'https://www.youtube.com/playlist?list=LL',
3429 ie=YoutubeTabIE.ie_key())
3430
3431
3432 class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
3433 IE_DESC = 'YouTube.com searches'
3434 # there doesn't appear to be a real limit, for example if you search for
3435 # 'python' you get more than 8.000.000 results
3436 _MAX_RESULTS = float('inf')
3437 IE_NAME = 'youtube:search'
3438 _SEARCH_KEY = 'ytsearch'
3439 _SEARCH_PARAMS = None
3440 _TESTS = []
3441
3442 def _entries(self, query, n):
3443 data = {
3444 'context': {
3445 'client': {
3446 'clientName': 'WEB',
3447 'clientVersion': '2.20201021.03.00',
3448 }
3449 },
3450 'query': query,
3451 }
3452 if self._SEARCH_PARAMS:
3453 data['params'] = self._SEARCH_PARAMS
3454 total = 0
3455 for page_num in itertools.count(1):
3456 search = self._download_json(
3457 'https://www.youtube.com/youtubei/v1/search?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
3458 video_id='query "%s"' % query,
3459 note='Downloading page %s' % page_num,
3460 errnote='Unable to download API page', fatal=False,
3461 data=json.dumps(data).encode('utf8'),
3462 headers={'content-type': 'application/json'})
3463 if not search:
3464 break
3465 slr_contents = try_get(
3466 search,
3467 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
3468 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
3469 list)
3470 if not slr_contents:
3471 break
3472
3473 isr_contents = []
3474 continuation_token = None
3475 # Youtube sometimes adds promoted content to searches,
3476 # changing the index location of videos and token.
3477 # So we search through all entries till we find them.
3478 for index, isr in enumerate(slr_contents):
3479 if not isr_contents:
3480 isr_contents = try_get(
3481 slr_contents,
3482 (lambda x: x[index]['itemSectionRenderer']['contents']),
3483 list)
3484 for content in isr_contents:
3485 if content.get('videoRenderer') is not None:
3486 break
3487 else:
3488 isr_contents = []
3489
3490 if continuation_token is None:
3491 continuation_token = try_get(
3492 slr_contents,
3493 lambda x: x[index]['continuationItemRenderer']['continuationEndpoint']['continuationCommand'][
3494 'token'],
3495 compat_str)
3496 if continuation_token is not None and isr_contents:
3497 break
3498
3499 if not isr_contents:
3500 break
3501 for content in isr_contents:
3502 if not isinstance(content, dict):
3503 continue
3504 video = content.get('videoRenderer')
3505 if not isinstance(video, dict):
3506 continue
3507 video_id = video.get('videoId')
3508 if not video_id:
3509 continue
3510 title = try_get(video, lambda x: x['title']['runs'][0]['text'], compat_str)
3511 description = try_get(video, lambda x: x['descriptionSnippet']['runs'][0]['text'], compat_str)
3512 duration = parse_duration(try_get(video, lambda x: x['lengthText']['simpleText'], compat_str))
3513 view_count_text = try_get(video, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
3514 view_count = int_or_none(self._search_regex(
3515 r'^(\d+)', re.sub(r'\s', '', view_count_text),
3516 'view count', default=None))
3517 uploader = try_get(video, lambda x: x['ownerText']['runs'][0]['text'], compat_str)
3518 total += 1
3519 yield {
3520 '_type': 'url_transparent',
3521 'ie_key': YoutubeIE.ie_key(),
3522 'id': video_id,
3523 'url': video_id,
3524 'title': title,
3525 'description': description,
3526 'duration': duration,
3527 'view_count': view_count,
3528 'uploader': uploader,
3529 }
3530 if total == n:
3531 return
3532 if not continuation_token:
3533 break
3534 data['continuation'] = continuation_token
3535
3536 def _get_n_results(self, query, n):
3537 """Get a specified number of results for a query"""
3538 return self.playlist_result(self._entries(query, n), query)
3539
3540
3541 class YoutubeSearchDateIE(YoutubeSearchIE):
3542 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3543 _SEARCH_KEY = 'ytsearchdate'
3544 IE_DESC = 'YouTube.com searches, newest videos first'
3545 _SEARCH_PARAMS = 'CAI%3D'
3546
3547
3548 class YoutubeSearchURLIE(YoutubeSearchIE):
3549 IE_DESC = 'YouTube.com search URLs'
3550 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
3551 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
3552 # _MAX_RESULTS = 100
3553 _TESTS = [{
3554 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3555 'playlist_mincount': 5,
3556 'info_dict': {
3557 'title': 'youtube-dl test video',
3558 }
3559 }, {
3560 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3561 'only_matching': True,
3562 }]
3563
3564 @classmethod
3565 def _make_valid_url(cls):
3566 return cls._VALID_URL
3567
3568 def _real_extract(self, url):
3569 qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
3570 query = (qs.get('search_query') or qs.get('q'))[0]
3571 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
3572 return self._get_n_results(query, self._MAX_RESULTS)
3573
3574
3575 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
3576 """
3577 Base class for feed extractors
3578 Subclasses must define the _FEED_NAME property.
3579 """
3580 _LOGIN_REQUIRED = True
3581 # _MAX_PAGES = 5
3582 _TESTS = []
3583
3584 @property
3585 def IE_NAME(self):
3586 return 'youtube:%s' % self._FEED_NAME
3587
3588 def _real_initialize(self):
3589 self._login()
3590
3591 def _real_extract(self, url):
3592 return self.url_result(
3593 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3594 ie=YoutubeTabIE.ie_key())
3595
3596
3597 class YoutubeWatchLaterIE(InfoExtractor):
3598 IE_NAME = 'youtube:watchlater'
3599 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3600 _VALID_URL = r':ytwatchlater'
3601 _TESTS = [{
3602 'url': ':ytwatchlater',
3603 'only_matching': True,
3604 }]
3605
3606 def _real_extract(self, url):
3607 return self.url_result(
3608 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
3609
3610
3611 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3612 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3613 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
3614 _FEED_NAME = 'recommended'
3615 _TESTS = [{
3616 'url': ':ytrec',
3617 'only_matching': True,
3618 }, {
3619 'url': ':ytrecommended',
3620 'only_matching': True,
3621 }, {
3622 'url': 'https://youtube.com',
3623 'only_matching': True,
3624 }]
3625
3626
3627 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3628 IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)'
3629 _VALID_URL = r':ytsub(?:scription)?s?'
3630 _FEED_NAME = 'subscriptions'
3631 _TESTS = [{
3632 'url': ':ytsubs',
3633 'only_matching': True,
3634 }, {
3635 'url': ':ytsubscriptions',
3636 'only_matching': True,
3637 }]
3638
3639
3640 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3641 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3642 _VALID_URL = r':ythistory'
3643 _FEED_NAME = 'history'
3644 _TESTS = [{
3645 'url': ':ythistory',
3646 'only_matching': True,
3647 }]
3648
3649
3650 class YoutubeTruncatedURLIE(InfoExtractor):
3651 IE_NAME = 'youtube:truncated_url'
3652 IE_DESC = False # Do not list
3653 _VALID_URL = r'''(?x)
3654 (?:https?://)?
3655 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3656 (?:watch\?(?:
3657 feature=[a-z_]+|
3658 annotation_id=annotation_[^&]+|
3659 x-yt-cl=[0-9]+|
3660 hl=[^&]*|
3661 t=[0-9]+
3662 )?
3663 |
3664 attribution_link\?a=[^&]+
3665 )
3666 $
3667 '''
3668
3669 _TESTS = [{
3670 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3671 'only_matching': True,
3672 }, {
3673 'url': 'https://www.youtube.com/watch?',
3674 'only_matching': True,
3675 }, {
3676 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3677 'only_matching': True,
3678 }, {
3679 'url': 'https://www.youtube.com/watch?feature=foo',
3680 'only_matching': True,
3681 }, {
3682 'url': 'https://www.youtube.com/watch?hl=en-GB',
3683 'only_matching': True,
3684 }, {
3685 'url': 'https://www.youtube.com/watch?t=2372',
3686 'only_matching': True,
3687 }]
3688
3689 def _real_extract(self, url):
3690 raise ExtractorError(
3691 'Did you forget to quote the URL? Remember that & is a meta '
3692 'character in most shells, so you want to put the URL in quotes, '
3693 'like youtube-dl '
3694 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3695 ' or simply youtube-dl BaW_jenozKc .',
3696 expected=True)
3697
3698
3699 class YoutubeTruncatedIDIE(InfoExtractor):
3700 IE_NAME = 'youtube:truncated_id'
3701 IE_DESC = False # Do not list
3702 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3703
3704 _TESTS = [{
3705 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3706 'only_matching': True,
3707 }]
3708
3709 def _real_extract(self, url):
3710 video_id = self._match_id(url)
3711 raise ExtractorError(
3712 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3713 expected=True)
3714
3715
3716 # Do Youtube show urls even exist anymore? I couldn't find any
3717 r'''
3718 class YoutubeShowIE(YoutubeTabIE):
3719 IE_DESC = 'YouTube.com (multi-season) shows'
3720 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3721 IE_NAME = 'youtube:show'
3722 _TESTS = [{
3723 'url': 'https://www.youtube.com/show/airdisasters',
3724 'playlist_mincount': 5,
3725 'info_dict': {
3726 'id': 'airdisasters',
3727 'title': 'Air Disasters',
3728 }
3729 }]
3730
3731 def _real_extract(self, url):
3732 playlist_id = self._match_id(url)
3733 return super(YoutubeShowIE, self)._real_extract(
3734 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3735 '''